[libc] Compile the GPU functions with '-fconvergent-functions' (#70229)

Summary:
This patch simply adds the `-fconvergent-functions` flag to the GPU
compilation. This is in relation to the behaviour of SIMT
architectures under divergence. With the flag, we assume every function
is convergent by default and rely on the compiler's divergence analysis
to transform it if possible.

Fixes: https://github.com/llvm/llvm-project/issues/63853