Add support for argument mutation in complex argument functions

[simon-id]

Closes #20

I think it doesn't work with class constructor rewrite ?

DISCLAIMER:

I have no idea what i'm doing, i'm only trying random stuff until the compiler and the tests are happy. Don't assume anything is correct.

[rochdev]

I think it doesn't work with class constructor rewrite ?

Shouldn't a test be failing if that's the case?

[simon-id]

nah there was no test for argument mutation, so i had to write one, and i only wrote it for function declaration, not for class constructor, i'll try to write one later to see if i'm right, but from what i can tell, the function that handles constructors doesn't go through my code path so i assume it doesn't work.
I already started coding a fix for it but I still have ZERO idea what i'm doing so idk

[timfish]

@bizob2828 asked me to review this PR on our temporary fork so I thought I should review this upstream since it is originally your PR!

[lforst]

@timfish @AbhiPrasad you guys think we can get this in?

[timfish]

I've approved it before so if the merge conflicts get fixed I don't see why not.

Did this already get merged in the fork? Need to check this out because I have no memory of reviewing this before!