[llvm] [RISCV][VLOPT] Add fsqrt instruction to isSupportInstr (PR #127462)

[Pengcheng Wang via llvm-commits]

================
@@ -1318,11 +1318,10 @@ define void @sqrt_v6bf16(ptr %x) {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 6, e16, m1, ta, ma
 ; CHECK-NEXT:    vle16.v v8, (a0)
-; CHECK-NEXT:    vsetivli zero, 8, e16, m1, ta, ma
----------------
wangpc-pp wrote:

I think the original CodeGen is wrong:
> A vector floating-point exception at any active floating-point element sets the standard FP exception flags in the fflags register. Inactive elements do not set FP exception flags.

As what we have discussed in https://github.com/llvm/llvm-project/pull/126483#discussion_r1948847405. The tail 2 elements may change `fflags` so the semantics is not the same as the LLVM IR.

https://github.com/llvm/llvm-project/pull/127462