E2e: Use pull through cache for container images

[lentzi90]

We risk hitting pull rate limits if we pull container images without authentication directly from the container registries. Luckily we have a Nordix provided Harbor instance that can be used as pull through cache for docker hub. We should configure the e2e tests to use this.

Relevant images:

• Images used when building the BMO image: https://github.com/metal3-io/baremetal-operator/blob/63abe2db90306baa805680ec391b055d9fe5be31/Dockerfile#L2-L3
• Ironic and cert-manager images: https://github.com/metal3-io/baremetal-operator/blob/63abe2db90306baa805680ec391b055d9fe5be31/test/e2e/config/ironic.yaml#L5-L13

We can use registry.nordix.org/docker-hub-proxy instead of docker.io and registry.nordix.org/quay-io-proxy instead of quay.io.

What to do:

1. Override the BASE_IMAGE and BUILD_IMAGEin ci-e2e.sh
2. Change the image BMO image used for older releases (release-0.5, release-0.4 and release-0.3) to use the pull through cache.

   images:
   - name: quay.io/metal3-io/baremetal-operator
    newName: registry.nordix.org/quay-io-proxy/metal3-io/baremetal-operator
    newTag: release-0.3

3. Update the ironic kustomization in the same way.
4. (Bonus, can be left for another PR) Work on the cert-manager installation procedure to make it possible to override the image. One option is to create a kustomization for it.

[lentzi90]

/triage accepted /cc @Rozzii

[metal3-io-bot]

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

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

/lifecycle stale

[lentzi90]

/remove-lifecycle stale

[metal3-io-bot]

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

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

/lifecycle stale

[Rozzii]

/remove-lifecycle stale /lifecycle frozen