[llvm] [AMDGPU][CodeGen][True16] Track waitcnt of vgpr32 instead of vgpr16 for 16bit reg in GFX11 (PR #157795)

[Joe Nash via llvm-commits]

================
@@ -586,6 +586,12 @@ def FeatureRealTrue16Insts : SubtargetFeature<"real-true16",
   "Use true 16-bit registers"
 >;
 
+def Feature16bitD16HWBug : SubtargetFeature<"d16-hw-bug",
+  "Enable16bitD16HWBug",
+  "true",
+  "D16 for 16 bit data type interfere the other half in true16 mode"
----------------
Sisyph wrote:

```suggestion
  "D16 instructions potentially have 32-bit data dependencies"
```


https://github.com/llvm/llvm-project/pull/157795