chore(ci): gather logs when bootstrap success or failed

[universal-itengineer]

Description

Ensure the reusable E2E pipeline always prepares and uploads bootstrap artifacts, even when dhctl-bootstrap fails before generating the full set of files.

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

When nested cluster bootstrap fails, the workflow used to skip the artifact preparation and upload steps because they only ran after a successful bootstrap. That made postmortem debugging harder: generated files, partial configuration, and related bootstrap outputs were not preserved in failed runs.

This change makes artifact preparation and upload run on both success and failure, and tolerates missing files that are not produced in failed bootstrap scenarios. As a result, failed runs still publish the available bootstrap data needed for investigation.

What is the expected result?

1. Run the reusable E2E pipeline.
2. If bootstrap succeeds, generated files are prepared and uploaded as before.
3. If bootstrap fails, the workflow still sanitizes and uploads whatever bootstrap files were produced.
4. Missing optional files do not break artifact preparation.

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: ci
type: chore
summary: Preserve generated bootstrap artifacts for reusable E2E pipeline runs even when bootstrap fails.
impact_level: low