Welcome.
1. There's no doubt that Alucard is possibly one of the most versatile video game characters with a wide range of moves to choose from whether it means traversing certain landscapes with transformations or fighting with specialized abilities.
Saying there's "no doubt" followed by "possibly" is a bit of contradiction. 
Banjo and Kazooie had alot of world-navigation abilities that you learn as you play.
The 2D Zelda games do that alot also, though via items, and it's alot more unnatural. ("Hey, I need an arrow to hit this weird eye switch! Hey, there's a bow and arrow in this dungeon! Hey, there's conviently hundreds of these eye switches in every other dungeon!"). Conceptually the same idea though.
Personally, I enjoy this kind of progression gating. It makes the worlds feel more open, while still letting designers keep certain areas off-limits until the appropriate time.
Another I like is having an area with really power enemies, so the player knows they aren't *supposed* to be there yet, but they feel even better when they are powerful enough to go back and conquer that area. Or, if they press forward at a low level anyway, they feel great when they are able to sneak-around or leverage items and tactics to 'defeat' enemies way over their level (as long as the game makes sure they can't accidentally trap themselves without a way back). This isn't a real solid progression gate though, since they can still get through. It's more of a "soft gate", if I might coin a term. The TV trope for this specific soft-gate is Beef Gate.
The game is built around your upgraded abilities and the likes in such a way that that you're able to progress to new areas with your new found abilities giving you a sense of progression and growth. What I want to know is if you think it'd be a good idea to start a Metroidvania off with a character already having most of that versatility and strength?
You're talking about metroidvanias... Some Metroid games do exactly that (especially some of the Metroid Primes - which may not be considered strictly metroidvanias, depending on your criteria). You start off with all (or most) your equipment at the beginning, and then lose them after about an hour of gameplay, so you still do progress afterward. I suppose this doesn't really count, because you can't explore the world yet, being trapped in the intro area.
If I remember correctly, Banjo and Tooie (sequel to Banjo and Kazooie) started off with you already having all the moves you had in the first game, so you had the full navigational ability (and comprehension on how to use it), while still gaining additional moves (and additional progression) as you play the game.
Mario 64 starts you have with your complete moveset (and gates you using the number of stars you collected from each world), which is weird when a level requires you to use a move that you've only had to use two or three times in the game (Wall Kicks Will Work!), but still works out fine. The sense of progression can occur using things other than world-transgressing abilities, like RPG-esqe level ups or equipment or health increases, or even just progressing through defeating each boss.
The sense of progression could be lost but with a versatile character right off the bat wouldn't be easier to bring some more sophisticated level design into the mix with that versatility in mind?
Yes, but at the cost of confusing inexperienced players, and even confusing experienced players who may get lost by following too many side-trails off the beaten path. If your game is designed to be fully open-world, or open-world with gates, it also changes the way you have to deliver the plot and balance the enemies, because you don't know whether the player will visit location X first or location Y, and the NPCs need to react to that, game events need to react to that, dungeon obstacles need to take that into account, and
One game that does both gating AND giving you the ability to go anywhere right off the bat, is the recently released indie game The Witness, because it uses knowledge in your own head, as the gate. It's a puzzle game, which may not be your cup of tea, but conceptually it's relevant to this discussion, because players CAN go solve any puzzle, but the puzzles are too complex, and so it's harder to do unless you first understand the individual elements of the puzzle learned from earlier puzzles. Basically, this is an knowledge-based "soft-gating" compared to the stat-based soft-gating I mentioned earlier.
So yes, you can definitely do that, and it makes for interesting game worlds. It does, however, create a different style of game, and a different way players have to play the game. If dropped in the middle of a world where you can go almost anywhere, it requires a different approach for players than being dropped in the middle of a world that guides you from location to location only gradually opening up the world.
It's a very fun difference, and one that I greatly enjoy. But it is a big difference to be aware of.
Basically, you're replacing hard-gates with soft-gates: Instead of requiring the player to get Ability A and making sure he knows how to use it, the obstacle is being gated by how well the player knows how to use it without being taught it as thoroughly. Further, if you give the player all the abilities at the start, he may forget which abilities he has, because he wasn't taught them one at a time.
2. We know for Megaman games that bosses often have a weakness that can be exploited, in most Megaman games said bosses are stunned when hit with that specific weakness. These powers that Megaman uses can also be applied to levels as well though some powers are 'hit' and most are usually a 'miss' when trying to give them a practical use. So my next question is, if we were to scrap the idea of a weakness that stun's or a weakness in general and make it more about how the player chooses to approach said boss or level with whatever ability they choose would that be a good design choice?
It's one way to do things that can be really great, but whether or not it's a "good" design choice depends on how well it plays with all the other design choices in a game.
It's like designing someone's living room. I can say, "Wow, this couch is nice.", "Yes, I like that desk.", "Man, this is a beautiful color.", "This other color is also very great.", "That table is fantastic!"
But all those awesome things may conflict with each other, and make an eclectic mess, instead of a beautiful room.
I see home-design objects only in isolation, I can't see how they fit together. Good home-decor designers either intuitively or by training, see all the pieces and how they fit together, just in their head. I've seen some designers making choices that I've actually tried to argue them out of, because I thought some of the individual components on their own looked horrendous, but they knew how it worked together, where in proper relation to other objects, even the horrendous objects looked wonderful.
Game design is the same way. You can look at individual mechanics or features or design elements and say, "That's awesome! It'll be great!", but when put together with other great elements, it can clash and do horribly. In addition, sometimes mechanics or design elements that sound terrible, work together with the rest of the design to become great.
Think Mario: "If you touch an enemy, or if an enemy touches you, you instantly die." Um, yay? But put together with all the other design elements of Mario, it works very very well.
Since it's not so much about exploitation as it is about how the player utilizes their abilities. Like a Link to the past, you have all these items but however you choose to fight the boss is up to you
The original Deus Ex and Thief were like that as well.
do you think this can be applied to a 2D action platformer?
Sure, if it works with all the other elements of the game's design.
3. My next question is about plot progression. In Megaman games you are able to choose from 8 levels right off the bat, you're not limited to playing in a specific order and you have the freedom to defeat the 8 bosses in whatever order you wish. The thing is, these bosses aren't essential to a plot and therefore don't effect the overarching story, they're simply a means to an end. And let's say we do make them essential to the plot, it'd be hard to tell a story and build upon it when you're choosing to fight these bosses in whatever order you wish. So for someone who wants to tell a good story, how do you circumvent this? Do we pull a Paper Mario and have at the end of every level a cut scene that furthers the story and develops the characters in some way? Do we scrap freedom of level choice altogether? Or do compromise and simply try our best to write the story with freedom of level choice in mind?
It is impossible to take story into account even given the player-driven flow of the encounters. You can do things like tweak up the dialog and change events based on which bosses the player has already defeated, for example.
Imagine two of the bosses a brother and sister in separate areas of the world. If you encounter the sister first, she might say something like, "I'll teach you to respect our master's will!", but if you meet her after killing her brother first, she might say, "You killed my brother! I'll make you suffer, and use your blood to revive him!", or whatever.
By having a hundred small details that tweak the game based on what the player has already done,
Zelda: Oracle of Ages and Oracle of Seasons tweaked up their worlds just slightly here and there, depending on whether you had already beaten on game and were carrying your save file over to the other. Things like encountering NPCs who you met in the other world, or certain mini-bosses replaced with antagonists from the other game looking for revenge, or antagonists from the other game unexpectedly attacking you in-between dungeons while travelling on the overworld. Further, it revealed subtle plot clues that if put together from both worlds, reveal more background plot, and help tie the two worlds together.
4. I already asked about versatility in Metroidvania's, but can that same versatility be applied to a 2D action platformer too? Especially in terms of fighting capability? Is it possible to make a 2D action platformer where the main character is capable of fighting like Bayonetta ( exaggerated example I know but it's just to get my point through )? Of course the levels and enemies would be designed around those fighting capabilities so do you think it's possible?
I haven't played Bayonetta, so I can't say for sure, but certainly you can implement the combat system you want, as long as the rest of the game takes it into account and works with it and is influenced from it.
The bigger question is how difficult would it be for you personally to implement it, whether the required skill levels are within your capabilities, and how much art and other content would it require, and whether all that content is within your budget.
It's not what's possible in general, but what's possible within what you have at your disposal. Necessity is the mother of invention. Limitation helps refine art and grow artists.
5. My final question is about flow. Do you think it's better to design for a video game to be designed around moving at a certain pace so that player never stops moving or design it in such a way that the player has to adjust their pace accordingly to problem solve by using their specific abilities to their advantage?
That's a very vague question, first entirely depending on the nature of the game's design, and secondarily depending on however you define "best". It sounds like there'd be pros and cons to both, which should influence, and should be influenced by, the rest of the design. It'd create an entirely different feel to the game.
