[clang] [Clang] Add `__builtin_experimental_vectorcompress` (PR #102476)

[Lawrence Benson via cfe-commits]

lawben wrote:

@philnik777 This PR is still waiting for #104904 and #105515 to be merged, so we have support for vaid x86, NEON, and SVE vectors. Once those PRs are in, this can probably be merged too.

https://github.com/llvm/llvm-project/pull/102476