Periodic job for CDI make rpm-deps

[cfilleke]

Is your feature request related to a problem? Please describe:
make rpm-deps is not a periodic job to be run for CDI as it is for kubevirt

Describe the solution you'd like:
In kubevirt there is a periodic job configured for make rpm-deps in /project-infra/github/ci/prow-deploy/files/jobs/kubevirt/kubevirt/kubevirt-periodics.yaml that updates the rpm dependencies weekly and reads:

- annotations:
    testgrid-create-test-group: "false"
  cluster: kubevirt-prow-control-plane
...
      - ../project-infra/hack/git-pr.sh -c "cd ../kubevirt && make rpm-deps" -b bump-rpm-dependencies

There is no automation for periodically running make rpm-deps for CDI, however.

The proposal here is to add a prow job to project-infra/github/ci/prow-deploy/files/jobs/kubevirt/containerized-data-importer/containerized-data-importer-periodics.yaml that executes make rpm-deps and opens the PR to update the containerized-data-importer/WORKSPACE and containerized-data-importer/rpm/BUILD.bazel files in main. @awels has suggested we make this a monthly periodic job.

Describe alternatives you've considered:
The alternative to running this out of a prow job is to run make rpm-deps manually for CDI and then create a PR that updates containerized-data-importer/WORKSPACE and containerized-data-importer/rpm/BUILD.bazel manually. This is how it is done now.

Additional context:
It's not clear what vetting of the dependent rpms is being conducted, e.g. for CVE issues and breaking functional changes. These things are encountered ad hoc.

The latest manual run of make rpm-deps PR is here: kubevirt/containerized-data-importer#3405

Issue initially raised in: https://groups.google.com/g/kubevirt-dev/c/pe__J0KT-zk/m/tx8-k54SCAAJ

[cfilleke]

When make rpm-deps is kicked off manually, there were some hiccups having to do with an older versions of nginx and https imports needing to be frozen in: kubevirt/containerized-data-importer#3405 (comment) so make sure those versions are frozen in future runs of make rpm-deps before automating the process in project-infra.

[kubevirt-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

/lifecycle stale