[llvm] [AMDGPU] Fix SIFoldOperandsImpl::tryFoldZeroHighBits when met non-reg src1 operand. (PR #133761)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,15 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
+# RUN: llc -mtriple=amdgcn -mcpu=gfx1101 -verify-machineinstrs -run-pass si-fold-operands %s -o - | FileCheck %s
+---
+name: test_tryFoldZeroHighBits_skips_nonreg
+tracksRegLiveness: true
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: test_tryFoldZeroHighBits_skips_nonreg
+    ; CHECK: [[V_MOV_B32_e32_:%[0-9]+]]:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
+    ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[V_MOV_B32_e32_]], %subreg.sub0, [[V_MOV_B32_e32_]], %subreg.sub1
+    ; CHECK-NEXT: [[V_AND_B32_e64_:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 65535, 0, implicit $exec
+  %0:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
+  %1:vreg_64 = REG_SEQUENCE %0, %subreg.sub0, %0, %subreg.sub1
+  %2:vgpr_32 = V_AND_B32_e64 65535, %1.sub0, implicit $exec
+
----------------
arsenm wrote:

```suggestion
  %2:vgpr_32 = V_AND_B32_e64 65535, %1.sub0, implicit $exec
  S_NOP 0, implicit %2
...
```

Missing end of function marker and better to avoid dead instructions

https://github.com/llvm/llvm-project/pull/133761