[llvm] [AMDGPU][CodeGen] enable D16Writes32BitVgpr for gfx12 (PR #165587)

[Brox Chen via llvm-commits]

broxigarchen wrote:

> There should be a test. For example we should see a new waitcnt on the true16 gfx12 path in some test.

It seems we don't have any gfx12 runline that generates a 16bit related waitcnt. Update one test runline to generate gfx12 runline with/without the d16writevgpr32 attribute. But still it's not very straightforward in the diff

https://github.com/llvm/llvm-project/pull/165587