[llvm] [LV] Improve materialisation of VFxUF for scalable vectors (PR #216708)

[David Sherwood via llvm-commits]

================

----------------
david-arm wrote:

I think you can also see from the IR changes that it looks right, e.g.

```
 ; CHECK:       [[VECTOR_PH]]:
 ; CHECK-NEXT:    [[TMP2:%.*]] = shl nuw i64 [[TMP0]], 3
-; CHECK-NEXT:    [[TMP3:%.*]] = shl nuw i64 [[TMP2]], 1
+; CHECK-NEXT:    [[TMP3:%.*]] = shl nuw i64 [[TMP0]], 4
```

- both before and after are shifting by 4 in total

https://github.com/llvm/llvm-project/pull/216708