Improve primitive biginteger operations

[jon-chuang]

Description

Details and benchmarks discussed here: #198
Related PR: #199, now #205

---

Before we can merge this PR, please make sure that all the following items have been
checked off. If any of the checklist items are not applicable, please leave them but
write a little note why.

• Targeted PR against correct branch (master)
• Linked to Github issue with discussion and accepted design OR have an explanation in the PR that describes this work.
• Wrote unit tests
• Updated relevant documentation in the code
• Added a relevant changelog entry to the Pending section in CHANGELOG.md
• Re-reviewed Files changed in the Github PR explorer

[weikengchen]

@Pratyush I think #205 and this one is solving the same issue and could be reconciled together.

[ValarDragon]

I think we need to benchmark this change on larger fields as well?

[jon-chuang]

Yes, you are probably right. This will currently involve changing a lot of Cargo.toml files to point to my local repo. I'll try to fix some of the issues I find in this process.

[jon-chuang]

Things I am not sure about:

• necessity of removing derive(Eq)?

[ValarDragon]

All of these changes LGTM. Holding off on approving until we have confirmed performance improvements of always unrolling on varying field sizes.

[Pratyush]

@ValarDragon this PR is subsumed by #205, minus the is_zero change (assembly is more compact for the iterator version at higher number of limbs, and at lower number of limbs it's unrolled). I adapted the rest of @jon-chuang's changes.

[ValarDragon]

I thought this PR also removed more asserts? I saw two removed here, but only one in the other PR. Let me recount though.

[ValarDragon]

Recounted, it is indeed not fully subsumed. The assert for mul_in_place removed here is not removed in #205 atm.

[jon-chuang]

closed in favour of #205