[X86] Prefer lowerVectorShuffleAsBitMask over using a avx512 masked operation when...

[X86] Prefer lowerVectorShuffleAsBitMask over using a avx512 masked operation when avx512bw/avx512vl is enabled.

This does require a constant pool load instead of loading an immediate into a gpr, moving to a k register and masking. But its less instructions and more consistent with previous ISAs. It probably opens up more combine opportunities as one of the test cases demonstrates.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@348018 91177308-0d34-0410-b5e6-96231b3b80d8