[llvm] [VPlan] First step towards VPlan cost modeling. (PR #92555)

[Florian Hahn via llvm-commits]

fhahn wrote:

Reverted for now in f1f3c34b4770437bdb022737918603b4bbeb523e. Would still be good to get the reproducer from @jeanPerier  for flang and from @dklimkin for tensorflow/XLA if possible. I believe they are distinct from the other issue mentioned here.

https://github.com/llvm/llvm-project/pull/92555