[llvm] [LV][POC] Use umin to avoid second-to-last iteration problems with EVL (PR #143434)

[Min-Yih Hsu via llvm-commits]

mshockwave wrote:

> If we do this at the IR level, I believe what you're describing _is_ the existing EVL implementation approach. If we don't change the IR, and do this as a late rewrite, then doing so in a sound manner is tricky.

I meant doing it (getting rid of minu instruction) later in the codegen pipeline, potentially after RISCVInsertVSETVLI, and wondering what should we do to ensure the soundness.

https://github.com/llvm/llvm-project/pull/143434