only create cachedir if requested, since otherwise opm will error out serving it

[grokspawn]

Description of the change:
Make cachedir creation contingent on a requested extractContent.cacheDir configuration.

Motivation for the change:
Otherwise, in-cluster opm will find an existing-but-empty cache directory in the filespace and will recognize it as a broken cache. This is correct, since it could be that the creator intended to put cache content in the destination, but forgot|failed to.

Architectural changes:

Testing remarks:
approach:

1. bring up a kind olm cluster via make run-local
2. clone the default operatorhubio catsrc but w/ on-cluster (cacheless) unpacking
   1. dump the catsrc yaml
   2. add extractConfig stanza (only configs)
   3. dump the catsrc contents opm render to local fs (in a subdir $catdir)
   4. opm generate dockerfile -i scratch $catdir to make a binless catsrc dockerfile
   5. edit the dockerfile & comment out the cache build/copy to the destination image
   6. build/push the image to an accessible registry (I used quay.io/jordankeister/catalogs:cacheless)
   7. change the modified catsrc yaml .status.image, ..metadata.name, spec.displayName to make this catsrc unique
   8. deploy the new catsrc via kubectl apply -f catsrc.yaml
3. profit!

Reviewer Checklist

• Implementation matches the proposed design, or proposal is updated to match implementation
• Sufficient unit test coverage
• Sufficient end-to-end test coverage
• Bug fixes are accompanied by regression test(s)
• e2e tests and flake fixes are accompanied evidence of flake testing, e.g. executing the test 100(0) times
• tech debt/todo is accompanied by issue link(s) in comments in the surrounding code
• Tests are comprehensible, e.g. Ginkgo DSL is being used appropriately
• Docs updated or added to /doc
• Commit messages sensible and descriptive
• Tests marked as [FLAKE] are truly flaky and have an issue
• Code is properly formatted