[JIT] Possible optimization for `x % {pow2_cns}` transformation for `ulong` types

[TIHan]

At the moment, we do not optimize x % {pow2_cns} to x & (pow2_cns - 1) for ulong types as they get transformed differently in morph.cpp. I believe we should only do this for X64 and not X86.

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch, @kunalspathak
See info in area-owners.md if you want to be subscribed.

Issue Details

---

At the moment, we do not optimize x % {pow2_cns} to x & (pow2_cns - 1) for ulong types as they get transformed differently in morph.cpp. I believe we should only do this for X64 and not X86.

Author:	TIHan
Assignees:	-
Labels:	tenet-performance, help wanted, area-CodeGen-coreclr
Milestone:	-