[sig-architecture] platform pods in ns/openshift-etcd should not exit an excessive amount of times

This new test appears to be a problem on etcd-scaling jobs where the exits are presently expected.

Example failure: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/periodic-ci-openshift-release-master-nightly-4.18-e2e-azure-ovn-etcd-scaling/1846579352007872512

An exception needs to be added, however we do not have a mechanism to add an exception for within a specific job right now, all we have to go on here is job name which is an imperfect way to disable these tests.

We don't want to disable the whole monitortest as that would shut down checks for pod exits on all the other control plane pods in the etcd-scaling job.

https://redhat-internal.slack.com/archives/C027U68LP/p1729174182218029 has details on why the test is presently expected to fail and some thoughts around how this could be solved:

via deads:
1. finding a way to avoid restarting containers is best
2. finding a way to make the exit more graceful is next best
3. skipping only the exact pod pattern on the exact test is next best (not on all jobs, only the scaling ones)

skipping the monitor test is not viable.

An option for 3 could be an env var that is applied only in the etcd-scaling job configuration, and the test could look for. Namespaces to skip the check on, comma separated or similar.