[MLIR][LLVM] Remove last remants of use-opaque-pointers from tests (#71076)

This commit removes the last remnants of `use-opaque-pointers` from the
mlir tests. Two of the tests seem to be disabled, while the CUDA one is
an integration test that didn't trigger a buildbot failure.

Author: Dinistro

mlir/test/Conversion/FuncToLLVM/convert-data-layout.mlir

@@ -1,6 +1,6 @@
 // RUN: mlir-opt -set-llvm-module-datalayout -convert-func-to-llvm %s | FileCheck %s
 
-// RUN-32: mlir-opt -set-llvm-module-datalayout='data-layout=p:32:32:32' -convert-func-to-llvm='use-opaque-pointers=1' %s \
+// RUN-32: mlir-opt -set-llvm-module-datalayout='data-layout=p:32:32:32' -convert-func-to-llvm %s \
 // RUN-32: | FileCheck %s
 
 // CHECK: module attributes {llvm.data_layout = ""}

mlir/test/Dialect/Linalg/roundtrip.mlir

@@ -4,7 +4,7 @@
 // TODO: Re-enable LLVM lowering test.
 //
 // Test that we can lower all the way to LLVM without crashing, don't check results here.
-// DISABLED: mlir-opt %s -='use-opaque-pointers=1' -o=/dev/null 2>&1
+// DISABLED: mlir-opt %s -o=/dev/null 2>&1
 
 func.func @views(%arg0: index) {
   %c0 = arith.constant 0 : index

mlir/test/Integration/GPU/CUDA/sm90/tma_load_64x8_8x128_noswizzle.mlir

@@ -22,7 +22,7 @@
 // RUN:         -convert-vector-to-llvm \
 // RUN:         -convert-index-to-llvm=index-bitwidth=32 \
 // RUN:         -convert-arith-to-llvm \
-// RUN:         -finalize-memref-to-llvm='use-opaque-pointers=1' \
+// RUN:         -finalize-memref-to-llvm \
 // RUN:         -convert-func-to-llvm \
 // RUN:         -expand-strided-metadata --nvvm-attach-target="module=main_kernel features=+ptx80 chip=sm_90 O=3" \
 // RUN:  | mlir-opt -pass-pipeline='builtin.module(gpu.module(strip-debuginfo,convert-gpu-to-nvvm,convert-index-to-llvm{index-bitwidth=32},canonicalize,cse))' \