[llvm] [LoadStoreVectorizer] Allow redundant stores (PR #169946)

[Gang Chen via llvm-commits]

cmc-rep wrote:

> Why? The vectorizer should not be making attempts to handle patterns that should have optimized away already

We have downstream needs for this: major IR transform happens after the last call of DSE, which creates new case of DSE.
We either need to add another DSE pass somewhere, or we add DSE-like capability here. The overhead of adding this on top of what we have added for redundant load is really minimal.

https://github.com/llvm/llvm-project/pull/169946