Description:
After the changes introduced in PR #524, the single/own namespace tests may start to flake because we are failing when the namespace or ClusterExtension is not deleted in time.

We also removed the SERIAL marker, so these tests now run in parallel.
This means that timing issues or delays in Kubernetes cleanup should not cause the test to fail.

Example failure:
See: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/periodic-ci-openshift-[…]ly-4.21-e2e-azure-ovn-runc-techpreview/1980106371709800448
 

      fail [/build/openshift/tests-extension/pkg/helpers/cluster_extension.go:185]: Timed out after 60.039s. Cleanup ClusterExtension install-webhook-bothns-ownns-ce-tz9c failed to delete Expected <bool>: false to be true

{{}}
Expected behaviour:
We should not fail the test if cleanup resources (like namespaces or ClusterExtensions) take longer to be deleted.
Kubernetes can take extra time to remove resources due to dependency cleanup, and that’s normal.

Proposed fix:

• Ensure test resources (namespaces, CEs, etc.) use unique names per scenario.

• Ensure we call Kubernetes delete for all resources properly.

• If deletion fails or takes too long, log a warning instead of failing the test — to avoid sending a false failure signal to Sippy or block other teams

Impact:
Unnecessary flakes and false failure signals on CI due to normal Kubernetes cleanup behaviour, which may block other teams and send bad signal to Sippy