[llvm] [Arch64][SVE] Lower svrev_* to llvm.vector.reverse and fold svrev(svrev(x)) -> x (PR #116422)

[Benjamin Maxwell via llvm-commits]

MacDue wrote:

> I'm still working on and trying out a few things, shouldn't have pushed. I need a little bit more time before a review, apologies.

You can set the PR as a draft to indicate it's WIP and then undraft it once you're ready for review. 

https://github.com/llvm/llvm-project/pull/116422