The width of the player's collision box makes jumping in narrow areas difficult, as does its height when trying not to bump your head on a corner and miss the jump you were aiming for.
The physics of accelerating isn't kind to jumping either, and makes platforming a little hard in general. Taking a moment to line up a jump is very counterproductive as when you jump while moving, the rate of acceleration is too low to reach many platforms that wouldn't be all that hard in most other games.

The enemy AI is pretty basic which is to be expected in early stages, it's just that when combined with how most enemies are just a tiny bit elevated the only tactic to beat them is jump and shoot, jump and shoot, repeat 'til dead. The AI isn't a problem at this point but yeah when they only spawn on ledges/platforms too small to share the space with and distance tactics are your only option, it's very same-y.

The aesthetic is great, and the plants and light/particle effects are beautiful. The enemies and protag only look out of place because they're not as pixel-rich as the rest of the world, which is likely on the to-do list to improve anyway. Procedurally-generated roguelite Metroidvania feels like a good time!

bump up the text speed in that prologue a TONNE, i was interested in what's being said but damn i got bored
then i clicked into a red/orange/yellow swirl thing and the game hung for like 30 seconds
sometimes when i try zapping to a corner i'll end up somwehere off the wall
no clue why i got sent into the void and called a fool
or why i'd spray black everywhere sometimes
or what i'm meant to do
i did get to the skrillex link but it sucked that i had to type it in myself, especially after typing a capital i instead of lowercase L
the music makes for a real interesting atmosphere tho, if i knew some more context it could be a pretty neat thing. some more things in the rooms other than 'big empty space' and 'enemies' might be nice too, like more things to jump off or something

walljumping felt unnecessarily difficult, when you jump it takes your vertical movement into account which meant as soon as you reached what would have been the apex of your jump into the wall and start sliding downwards, you don't get any upwards movement from any walljumps you attempt. that combined with needing to hold the direction the wall is in to walljump again made it infuriating.
bouncing up the walls felt cool the one time i managed to do it but it was such a pain to get there reflex- and timing-wise i'd had it. also, the enemies in the second area take too many shots to kill and i just kept dying there. might be the responsiveness of my mouse but it was still a pain.

similar to what one of the other comments said, try to figure out a theme, message, concept or 'point' for your game before you get much further in development. if you don't know what you're making (in the deeper sense than 'a platformer'), you aren't really making anything.

The pause at the end of each level is odd but should be an easy fix.
The graphics feel too close to the old 2000's flash game aesthetic which isn't necessarily bad but it did make it hard for me to take the game seriously at first. A lot of images have a bit of noise about them which makes them look just a bit off.
As far as puzzle elements go, the mirrors and quantum relocators were great tools, though the dark matter and whatever randomised the laser's direction felt more like frustrations than actual obstacles or tools. The dark matter just made it more important to aim very particularly, which could have been done with regular wall blocks instead of including an energy system (which doesn't seem to come into play very much, and for a thinking puzzle might be unecessary), and 'hoping for the best' is counter-productive for a puzzle game. The level it was introduced in was literally a matter of pointing the laser and waiting for the puzzle to accidentally solve itself, which isn't fun or rewarding.

The auto-spinning mirror was a nice touch, ie something that can't be used by the player but worked with/around, but again it was a case of letting the puzzle solve itself. If the bubbles in the laser's line of sight were all popped immediately it wouldn't feel this way, but since the laser could only reach one ball at a time (ie. it needing a few frames to burst, allowing the laser to have moved by the time the next one isn't obscured) it turned into a waiting game - particularly because the mirror I was using directly blocked the scientist from any danger and I didn't have to think about anything while waiting for the mirror to rotate again and again and again.

Puzzles becoming waiting games, ie necessitating you do nothing while the laser goes, makes the energy bar even less necessary since even waiting as long as I did for the mirrors to rotate or luck to be in my favour, the energy bar never depleted. In later levels this might be different, and conserving energy might turn into an actual element of the puzzles, but in every level I played it didn't matter at all.
Consider not showing/using the bar at all in levels that don't involve an element of conservation to solve, as while it isn't it looks like a meaningless addition - and, if conservation isn't part of the puzzles at any point, it just is.

Thanks very much, I really appreciate the feedback.
Would you mind elaborating on the issues with the physics? Is it the movement, the camera, what's messed up that gotta be unmessed?