test: add postcleanup mode

[yaroslavborbat]

Description

Replace boolean IsCleanupNeeded flag with PostCleanupMode enum in e2e test config.

New values for POST_CLEANUP env:

• always (default) — always clean up resources after tests
• never — never clean up (useful for debugging)
• no-failure — skip cleanup only when the test suite fails, preserving resources on failure for investigation (cleanup runs on success)

Why do we need it, and what problem does it solve?

The old boolean flag (yes/no) didn't allow keeping resources when tests fail while still cleaning up on success. The new no-failure mode fills this gap, making it easier to debug failing e2e tests without manually toggling cleanup.

What is the expected result?

1. Set POST_CLEANUP=no-failure before running e2e tests
2. If tests pass — resources are cleaned up as usual
3. If tests fail — resources remain in the cluster for investigation

Checklist

• The code is covered by unit tests.
• e2e tests passed.
• Documentation updated according to the changes.
• Changes were tested in the Kubernetes cluster manually.

Changelog entries

section: test
type: chore
summary: Replace boolean POST_CLEANUP flag with PostCleanupMode enum (always/never/no-failure) in e2e config.
impact_level: low

[hardcoretime]

Suggested change

	// Valid values: "always", "never", "no-failure", or "" (default = always).
	// Valid values: "always", "never", "no-on-failure", or "" (default = always).