[llvm] [LV, VP]VP intrinsics support for the Loop Vectorizer + adding new tail-folding mode using EVL. (PR #76172)

[via llvm-commits]

================
@@ -1633,6 +1634,23 @@ class LoopVectorizationCostModel {
     return foldTailByMasking() || Legal->blockNeedsPredication(BB);
   }
 
+  /// Returns true if VP intrinsics with explicit vector length support should
+  /// be generated in the tail folded loop.
----------------
ayalz wrote:

Why/is useVPIWithVPEVLVectorization() needed? Could getTailFoldingStyle() take care of checking all considerations before deciding on DataWithEVL, as done for other options?

https://github.com/llvm/llvm-project/pull/76172