E2E test infrastructure: migrate console plugin optimisations to backstage plugin

[Anton-Fil]

Summary

The console plugin (kuadrant-console-plugin) has accumulated several e2e CI/CD optimisations over the past months. This ticket tracks migrating the applicable ones to the backstage plugin.

What we have in console plugin

Feature	Description	Effort
Parallel workers	fullyParallel: true, workers: 3 + unique resource names per test	Medium — requires solving DEX login race conditions
@smoke / @nightly tags	Tests tagged by priority; PRs run only @smoke, nightly runs all	Low
Suite-router	Bash script mapping changed files → relevant spec files	Medium
Two-list routing	Component changes → --grep @smoke; test file changes → all tags	Low (built on top of suite-router)
Spec map validation	CI check that every spec file is mapped in suite-router	Low
Nightly workflow + auto-issue	Cron at 2:00 UTC, auto-creates GitHub issue on failure with failed test names	Low
Failure classification	Distinguishes infrastructure failures (cluster/server) from test failures	Low

Recommendation

Not everything makes sense right now — there are only 7 spec files and workers: 1 due to DEX popup login race conditions.

Worth doing now (low effort, immediate value):

• Nightly workflow with auto-issue creation
• Failure classification (infrastructure vs test)
• @smoke / @nightly tags (prepares for future suite-router)

Worth doing later (when test suite grows):

• Parallel workers (needs auth refactor — shared session or API-based login)
• Suite-router + spec map validation
• Two-list routing

Note

This needs discussion with the team before implementation — the backstage plugin has different constraints (DEX auth, kind cluster setup, smaller test suite) and some optimisations may not be worth the effort yet.