Leverage CI cache to bootstrap the test `test_node_with_rewriter`

Regarding the rewriting taking a long time, we might be able to set up some sort of cache-checking (do not attempt to rewrite if the rewritten binary already exists and the sha256sum of the input matches the checksum we've stored beside the rewritten binary). This would work pretty well on the CI _without_ checking large binaries in, because the CI _already_ has a caching step for anything that sits inside the `target/` directory (well technically, it caches a subset of even `target/` but we can tweak that quite easily). Anyways, I think that is an orthogonal concern, and need not block merging this PR in. Thanks again!

_Originally posted by @jaybosamiya-ms in https://github.com/microsoft/litebox/pull/300#pullrequestreview-3198496536_

Start with the test_node_with_rewriter test and inspect how rewriting outputs are placed under target/. Review the CI cache configuration and the discussion in pull request 300. Done means the test can reuse an existing rewritten binary when its stored input checksum still matches, without checking large binaries into the repository.

Written by the indexing model from the issue text.