[clang] [clang-tools-extra] [CudaSPIRV] Allow using integral non-type template parameters as attribute args (PR #131546)

[Matt Arsenault via cfe-commits]

================
@@ -0,0 +1,34 @@
+// RUN: %clang_cc1 -triple spirv64 -aux-triple x86_64-unknown-linux-gnu \
+// RUN:   -fcuda-is-device -verify -fsyntax-only %s
+
+#include "Inputs/cuda.h"
----------------
arsenm wrote:

Probably should just locally define __global__ instead of depending on the full input header 

https://github.com/llvm/llvm-project/pull/131546