Currently cerberus runs as a post step after workload is triggered, we should run it as observer in CI so that any failures for netobserv related pods can be caught as they occur while the workload is being run.

Example:
https://github.com/openshift/release/blob/7202c0f7e30362d470cacc51aa8a1abb8fc266c8/ci-operator/config/redhat-chaos/prow-scripts/redhat-chaos-prow-scripts-main__cr-4.20-nightly.yaml#L53-L55

To support for NetObserv more changes will be need to watch for netobserv related namespaces to existing step https://github.com/memodi/release/blob/master/ci-operator/step-registry/redhat-chaos/cerberus/redhat-chaos-cerberus-commands.sh