[llvm] AMDGPU: Handle legal v2f16/v2bf16 atomicrmw fadd for global/flat (PR #95394)

[Jay Foad via llvm-commits]

================
@@ -15942,6 +15942,16 @@ static bool isHalf2OrBFloat2(Type *Ty) {
   return false;
 }
 
+static bool isHalf2(Type *Ty) {
----------------
jayfoad wrote:

Can we make this `isV2F16` please? LLVM already has a couple of conventions for naming vector types, I don't think we need to add another one.

https://github.com/llvm/llvm-project/pull/95394