[llvm] [LV] Vectorize predicated loop-invariant loads via masked load + broadcast (PR #200361)

[Ramkumar Ramachandra via llvm-commits]

================
@@ -13,6 +13,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "LoopVectorizationPlanner.h"
+#include "VPlanTransforms.h"
----------------
artagnon wrote:

Unnecessary?

https://github.com/llvm/llvm-project/pull/200361