[llvm] [NFC][AMDGPU] Use m_Intrinsic and m_AnyIntrinsic PatternMatch combinators across the backend (PR #194337)

[Arseniy Obolenskiy via llvm-commits]

================
@@ -964,6 +964,7 @@ AMDGPUPromoteAllocaImpl::getVectorTypeForAlloca(Type *AllocaTy) const {
 }
 
 void AMDGPUPromoteAllocaImpl::analyzePromoteToVector(AllocaAnalysis &AA) const {
+  using namespace PatternMatch;
----------------
aobolensk wrote:

Done

https://github.com/llvm/llvm-project/pull/194337