[clang] [CUDA][Win32] Add `fma(long double,..)` to math forward declares. (PR #73756)

[Artem Belevich via cfe-commits]

================
@@ -70,6 +70,9 @@ __DEVICE__ double floor(double);
 __DEVICE__ float floor(float);
 __DEVICE__ double fma(double, double, double);
 __DEVICE__ float fma(float, float, float);
+#ifdef _MSC_VER
+__DEVICE__ long double fma(long double, long double, long double);
----------------
Artem-B wrote:

We already have a handful of MSVC-specific no-implementation functions declared here, so one more is OK.
I just want to document it better.

You may want to add a macro with a descriptive name. (E.g. `CUDA_ALLOW_LONG_DOUBLE_MATH_FUNCTION_DECLS`) and have it defined for MSVC, along with the comment why it's needed. Then replace the existing `#if _MSC_VER` with it.

https://github.com/llvm/llvm-project/pull/73756