[llvm] [AMDGPU][gfx1250] Also add a wait on xcnt before volatile accesses (PR #168852)

[Pierre van Houtryve via llvm-commits]

Pierre-vh wrote:

My understanding is that it's for any access that can/could go beyond the vector L0 cache (WGP$), so this is enabled for isFLAT (which includes `flat_`, `global_` and `scratch_`).
LDS cannot go above WGP$, and we don't have smem writes, so neither of these are affected.

https://github.com/llvm/llvm-project/pull/168852