Saving/restoring XMM / YMM register state in inline assembly

[pgoodman]

I don't recall if we're actually saving/restoring enough. I think we're only saving xmm0 through xmm7. I'd need to think more about whether more is needed. I don't believe that we rewrite things like raw uses of xmm8 (in avx2) to be in that range, for instance.

[pgoodman]

A better approach to saving/restoring state would be: instead of saving/restoring all registers, emit additional instructions for saving/restoring the used registers.

[pgoodman]

@awesie Feel free to tackle this :-) If you're looking for an idea on how to do it, it would be similar-ish to how we "rewrite" uses of esp/rsp to some other register.