Add debug info about failures in CI system using the kubernetes-namespace-report action

[consideRatio]

I quickly looked at a recent test failure following the merge of #185. For things erroring in tests done in a k8s cluster, it could be relevant to make use of this github action: https://github.com/jupyterhub/action-k8s-namespace-report

I'm a maintainer of the action and we are using it across the jupyterhub org on github.

This is how we use it:

      # ref: https://github.com/jupyterhub/action-k8s-namespace-report
      - name: Kubernetes namespace report
        uses: jupyterhub/action-k8s-namespace-report@v1
        if: always()
        with:
          important-workloads: deploy/hub deploy/proxy

This is how it looks, where it will help provide logs of relevance when it detects something that has errored, or always emit logs by having a workload declared under "important-workloads".