fix: avoid e2e teardown panic after setup failure

[immanuwell]

ShutdownOperatorProcess() can panic if suite startup exits before cancelManager is set. Then the real setup error gets buried, kinda noisy.

This just guards the "manager never started" path and adds a small regression test.

Repro

1. Point kubeconfig at an unreachable or not ready cluster.
2. Run go test ./test/e2e -run TestE2E.
3. Before this patch, SynchronizedBeforeSuite fails and SynchronizedAfterSuite panics on nil cancelManager.
4. After this patch, you only get the original setup failure.

---

Summary by cubic

Prevent e2e teardown panics when setup fails by safely handling cases where the operator manager never started. Teardown now skips manager shutdown and surfaces the original setup error.

• Bug Fixes
  • Guard manager shutdown: only call cancel and wait if the manager was initialized.
  • Add a regression test to ensure teardown does not panic when the manager never started.

^{Written for commit 3462f93. Summary will update on new commits.}

[cubic-dev-ai]

No issues found across 2 files

_{Re-trigger cubic}

[AndrewChubatiuk]

LGTM! Thanks for contribution