[llvm] [LV] Use wide lane masks as the canonical form when tail-folding & interleaving (PR #209484)

[Benjamin Maxwell via llvm-commits]

================
@@ -1232,6 +1232,10 @@ class LLVM_ABI_FOR_TEST VPInstruction : public VPRecipeWithIRFlags,
     //    mask[i] = icmpt ult (op0 + i), op1
     // The size of the mask returned is VF * Multiplier (UF, third op).
     ActiveLaneMask,
+    ActiveLaneMaskForControlFlow,
+    // Extracts each unrolled part of a widened active lane mask when used
+    // for control flow.
----------------
MacDue wrote:

nit: Maybe phrase this more generally.

```suggestion
    // Extracts each unrolled part of a (VFxUF) widened vector/mask.
```

https://github.com/llvm/llvm-project/pull/209484