test: group e2e dumps by testcase

[yaroslavborbat]

Description

Group e2e failure dumps by test case and simplify artifact filenames

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

Failed e2e artifacts are currently dumped into one flat directory with very long filenames that duplicate the full test name in every file. This makes diagnostics hard to navigate and noisy in CI artifacts.

This change stores each failed test dump in its own directory (<tmp>/<formatted-test-name>/) and writes files with concise, stable names inside that directory.

Examples of the new layout:

• resources.yaml
• pod_<pod>__<container>_logs.json
• pod_<pod>_describe.log
• intvirtvm_describe.log
• intvirtvmi_describe.log
• nodes_describe.log
• nodes_owide.log
• events_<namespace>.yaml
• clusternetwork.yaml, cep.yaml (for additional network tests)

Result: artifacts are easier to browse, compare, and attach to incident/debug workflows.

What is the expected result?

1. Run e2e tests with at least one failed case.
2. Open the artifacts directory (RUNNER_TEMP or /tmp).
3. Verify that each failed test has its own folder named by formatted test full text.
4. Verify that files inside each folder use short names (without repeated e2e_failed__<test-name>__... prefix).
5. Confirm that all previously collected dump categories are still present.

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.