[mlir] check for failures when packing function sigunatures in std->llvm conversion

[ftynse]

Summary

When packing function results into a structure during the standard-to-llvm
dialect conversion, do not assume the conversion was successful and propagate
nullptr as error state.

Fixes PR45184.

[ftynse]

Rebase

mlir/lib/Conversion/StandardToLLVM/StandardToLLVM.cpp

Type LLVMTypeConverter::packFunctionResults(ArrayRef<Type> types) {
   assert(!types.empty() && "expected non-empty list of type");
 
   if (types.size() == 1)
     return convertType(types.front());
 
   SmallVector<LLVM::LLVMType, 8> resultTypes;
   resultTypes.reserve(types.size());
   for (auto t : types) {
-    auto converted = convertType(t).dyn_cast<LLVM::LLVMType>();
+    auto converted = convertType(t).dyn_cast_or_null<LLVM::LLVMType>();
     if (!converted)
       return {};
     resultTypes.push_back(converted);
   }
 
   return LLVM::LLVMType::getStructTy(&getContext(), resultTypes);
 }
 

mlir/test/Conversion/StandardToLLVM/invalid.mlir

func @mlir_cast_to_llvm_vec(%0 : vector<1x1xf32>) -> !llvm.vec<1 x float> {
   %1 = llvm.mlir.cast %0 : vector<1x1xf32> to !llvm.vec<1 x float>
   return %1 : !llvm.vec<1 x float>
 }
 
 // -----
 
 // Should not crash on unsupported types in function signatures.
 func @unsupported_signature() -> tensor<10 x i32>
+
+// -----
+
+func @partially_supported_signature() -> (vector<10 x i32>, tensor<10 x i32>)