I completely forgot to mention it in my original post, but I think your overall level and world design was very solid. I never felt lost for particularly long, and some of the platforming challenges took a bit of thinking for my approach, but I think that's actually a good thing, it was fun to plan out where to jump, where to dash, etc. I definitely don't think level design is at fault for some of my frustrations, those came primarily from the high dexterity requirements. Perhaps with some fixes on the controls, the execution barrier wouldn't have felt as high.
I don't feel qualified to say what would be "better" for the game. After all, Celeste is an excellent game, but I don't think it needs a combat system and boss fights (it does kind of have boss fights, but they're really obstacle courses). If that's the kind of game you want to make, I think it could be very cool! I think this game could work with more of a focus on rhythm-based combat and exploration, or with a focus on precision platforming, depending on what you want to make. Perhaps it is possible to combine both, but I do think that'd alienate a larger portion of players, since those two are pretty different appeals. Don't let that discourage you from making that game if you really want to, though, perhaps there's potential! If you do want to keep developing this game, I'd recommend doing some prototyping of different approaches, and really honing on where the fun is, in your opinion.