feat(e2e): support running e2e tests on real OCP clusters with Prometheus alert validation

[rlobillo]

Short Description

Enable running e2e tests against real OCP clusters (not just Kind) and add Prometheus alert tests for VirtPlatformSyncFailed and VirtPlatformDependencyMissing.

More details

The existing e2e suite was designed for Kind-only. This PR adapts the test lifecycle to be non-destructive on real clusters and adds a new make run-e2e-tests-only target that runs Ginkgo directly against whatever cluster KUBECONFIG points to, producing JUnit XML and JSON reports in _output/.

Two new Prometheus alert tests are added:

• VirtPlatformSyncFailed (table-driven): creates a ValidatingWebhookConfiguration that blocks SSA Create and Update for each managed asset, then triggers reconciliation so the SSA dry-run fails and the CNV-89450 fix sets compliance_status=0. Asserts the alert fires with correct labels (kind, name, severity: critical, operator: virt-platform-autopilot). Assets whose CRD or gateCRD are missing are skipped automatically.
• VirtPlatformDependencyMissing (passive): queries the /metrics endpoint for kubevirt_autopilot_missing_dependency == 1, then verifies a warning alert fires for each missing optional CRD individually.

What this PR does / why we need it

Commit 1: feat(e2e): support running e2e tests against real OCP clusters

• Add make run-e2e-tests-only target (runs Ginkgo with JUnit/JSON reports)
• Adapt drift_test.go and anti_thrashing_e2e_test.go to use ensureHCOExists() / patchAutopilotAndWait() instead of deleting/recreating HCO (non-destructive on real clusters)
• crd_lifecycle_test.go skips on OCP (CRD creation/deletion only meaningful on Kind)
• Add isOpenShiftCluster() helper (detects OCP via ClusterVersion CRD presence)

Commit 2: feat(e2e): add Prometheus alert tests for OCP

• Table-driven VirtPlatformSyncFailed tests for all phase-1 assets (MachineConfig, KubeletConfig, NodeHealthCheck, UIPlugin, KubeDescheduler)
• queryFiringAlert() returns map[string]string labels with PromQL label filters — concise logs ("not firing yet" / "firing — kind=X name=Y severity=Z")
• queryMetricExists() with clean logs ("not found yet" / "found (N series)")
• touchHCO() triggers reconciliation before metrics wait (handles idle pods with no recent metrics)
• getMissingDependenciesFromMetrics() parses /metrics for individual missing dependencies
• Prometheus querying via thanos-querier route with SA token authentication
• webhookCreated flag guards AfterEach cleanup (prevents timeouts on skipped/failed tests)
• PrometheusRule for durations reduced to 15s for faster test feedback
• Metrics wait timeout: 10min (handles post-MCO-rollout scenarios, CNV-89454)

Commit 3: fix(e2e): remove unused namespace param from captureAssetMetrics

• Fix unparam lint: captureAssetMetrics namespace parameter always received operatorNamespace — use the package-level variable directly

Commit 4: fix(e2e): block Update ops in webhook to trigger SSA dry-run failure

• The blocking webhook only intercepted Create operations, but SSA dry-run uses Patch (Update). With the CNV-89450 fix now setting compliance_status=0 on dry-run failure, the webhook must also block Update so DetectDrift() errors out and the VirtPlatformSyncFailed alert fires.
• Removes the old workaround of deleting the resource to bypass dry-run drift detection.

🤖 Generated with Claude Code

[openshift-ci]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign tiraboschi for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[github-actions]

❌ Generated Files Verification Failed

One or more generated files in this PR are out of sync:

• CRDs: Run make update-crds if CRD verification failed
• RBAC: Run make generate-rbac if RBAC verification failed

Please regenerate the files locally and commit the changes.

[github-actions]

❌ Generated Files Verification Failed

One or more generated files in this PR are out of sync:

• CRDs: Run make update-crds if CRD verification failed
• RBAC: Run make generate-rbac if RBAC verification failed

Please regenerate the files locally and commit the changes.

[github-actions]

❌ Generated Files Verification Failed

One or more generated files in this PR are out of sync:

• CRDs: Run make update-crds if CRD verification failed
• RBAC: Run make generate-rbac if RBAC verification failed

Please regenerate the files locally and commit the changes.

[rlobillo]

/hold

Tests need to be updated as soon as #207 is ready on a build.

[github-actions]

❌ Generated Files Verification Failed

One or more generated files in this PR are out of sync:

• CRDs: Run make update-crds if CRD verification failed
• RBAC: Run make generate-rbac if RBAC verification failed

Please regenerate the files locally and commit the changes.

[rlobillo]

All tests passing on real OCP cluster

$ tail e2e.log 
autogenerated by Ginkgo
[ReportAfterSuite] PASSED [0.013 seconds]
------------------------------

Ran 17 of 27 Specs in 325.317 seconds
SUCCESS! -- 17 Passed | 0 Failed | 0 Pending | 10 Skipped
PASS

Ginkgo ran 1 suite in 5m27.191502866s
Test Suite Passed