[clang] [AMDGPU] Change -cl-std to CL2.0 in builtins-amdgcn-gfx950.cl. NFC. (PR #138745)

[Stanislav Mekhanoshin via cfe-commits]

rampitec wrote:

> > > Why? This just adds more intermediate IR and makes the output bigger. If anything we should switch any tests unnecessarily using CL2.0 to use CL1.2
> > 
> > 
> > It is like that everywhere, and that eliminates huge delta to downstream branch.
> 
> Fix the downstream branch

Ok.

https://github.com/llvm/llvm-project/pull/138745