[llvm] [RISCV] Add +optimized-nf{3, 4}-segment-load-store (PR #114414)

[Luke Lau via llvm-commits]

================
@@ -1386,6 +1386,13 @@ def TuneOptimizedZeroStrideLoad
                       "true", "Optimized (perform fewer memory operations)"
                       "zero-stride vector load">;
 
+foreach nf = {3-4} in
----------------
lukel97 wrote:

No particular reason, happy to add those in. 

Another related thought: Should we also have a feature for NF=2 and enable it by default in the generic tuning? That way processors could opt out of it too.

https://github.com/llvm/llvm-project/pull/114414