[llvm] [LV] Convert gather loads with invariant stride into strided loads (PR #147297)

[Pengcheng Wang via llvm-commits]

wangpc-pp wrote:

> I know the spacemit-x60 doesn't have fast vlse but I don't think they're as slow as vluxei. Do we have benchmark results for these? cc) @mikhailramalho

I have some data on C908 for strided load (which is weird): https://github.com/camel-cdr/rvv-bench/issues/12. The result may hold for spacemit-x60.

https://github.com/llvm/llvm-project/pull/147297