MSM: Bring in quotient polynomial and constraints

[volhovm]

Solves #1790:

1. Moves constraint creating logic into MSM prover/verifier
2. Hopefully wiring it up with extra simple circuit to prove it works

[dannywillems]

Closing. See serialization subdirectory (wrong PR, sorry)

[dannywillems]

Oopsie. Sorry, wrong PR.

[dannywillems]

@volhovm Can you fix the conflicts please?

[volhovm]

@dannywillems ah, oh well. on it.

[dannywillems]

Can you squash: be3d64e and bcd09ef please? They are unecessary commits.

[volhovm]

@dannywillems I know it's annoying but I can't really squash these without messing everything up sadly because I merged master into it, so now git rebase -i treats my commits and master commits equally. If you know a way to do it please tell me

[dannywillems]

@dannywillems I know it's annoying but I can't really squash these without messing everything up sadly because I merged master into it, so now git rebase -i treats my commits and master commits equally. If you know a way to do it please tell me

The only way I think is restarting the branch before the git merge, squash, redo git merge, git push force.

[volhovm]

@dannywillems all addressed

[dannywillems]

Let's just merge it. We'll fix later. We will iterate over it. I didn't test it, but we will after. It does create tons of conflicts in the next PRs.
Let's have very small PR later.

I would revert the changes in kimchi to decrease as much as possible the changes outside of the MSM code