Typo and potential null dereference in HLMatrixBitcastLowerPass.cpp

https://github.com/microsoft/DirectXShaderCompiler/blob/8f2881e51178346547c2bf58c74691e05c5cf825/lib/HLSL/HLMatrixBitcastLowerPass.cpp#L229C28-L229C32

The LI below should be ST. The LI is from the previous if testing for U's type.

    } else if (LoadInst *LI = dyn_cast<LoadInst>(U)) {
    ...
    } else if (StoreInst *ST = dyn_cast<StoreInst>(U)) {
      Value *V = ST->getValueOperand();
      if (VectorType *Ty = dyn_cast<VectorType>(V->getType())) {
        IRBuilder<> Builder(LI); // <---- Should be `ST`
        Value *zeroIdx = Builder.getInt32(0);
        unsigned vecSize = Ty->getNumElements();
        for (unsigned i = 0; i < vecSize; i++) {
          Value *GEP = CreateEltGEP(A, i, zeroIdx, Builder);
          Value *Elt = Builder.CreateExtractElement(V, i);
          Builder.CreateStore(Elt, GEP);
        }
        ST->eraseFromParent();

I made an issue instead of a PR because I don't have a test case, just noticed this typo.

Start at lib/HLSL/HLMatrixBitcastLowerPass.cpp around line 229 and compare the load and store branches shown in the issue. Confirm that the store branch uses ST when constructing IRBuilder, then build or validate the compiler as appropriate; the issue provides no test case, so done means the typo and potential null dereference are corrected without changing the surrounding behavior.