[llvm] [RISCV] Fold vector shift of sext/zext to widening multiply (PR #121563)

[Luke Lau via llvm-commits]

https://github.com/lukel97 commented:

Makes sense to me. I guess multiplies might not have the same number of available execution ports, but it's still less vector ops at the end of the day. 

https://github.com/llvm/llvm-project/pull/121563