[llvm] [AMDGPU] Add OMOD folding for TRANS bfloat16 instructions (PR #208595)

[Shilei Tian via llvm-commits]

================
@@ -2363,6 +2358,117 @@ SIFoldOperandsImpl::isOMod(const MachineInstr &MI) const {
 
     return std::pair(nullptr, SIOutMods::NONE);
   }
+  case AMDGPU::V_FMA_MIXLO_BF16:
+  case AMDGPU::V_FMA_MIX_BF16_t16: {
+    if (MFI->getMode().FP32Denormals.Output != DenormalMode::PreserveSign ||
+        MI.mayRaiseFPException())
+      return std::pair(nullptr, SIOutMods::NONE);
----------------
shiltian wrote:

```suggestion
      return {nullptr, SIOutMods::NONE};
```

as well as all similar uses afterwards.

https://github.com/llvm/llvm-project/pull/208595