[llvm] [RISCV] Support vectorizing FMINIMUMNUM and FMAXIMUMNUM (PR #135727)

[YunQiang Su via llvm-commits]

================
@@ -0,0 +1,201 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
----------------
wzssyqa wrote:

I  have double checking. My test seems to have better test coverage.
I have none 2^n length tests such <3 x float>.

I guess you mean something like `nnan`? It is not needed at all for fmaximumnum on RISC-V.

https://github.com/llvm/llvm-project/pull/135727