fix: serialize e2e workflow runs and harden repo orchestration

[arjuncooliitr]

This PR addresses the remaining failures in the app-test.yml workflow after disabling aio-lib-events and aio-lib-web.

Root Cause

The workflow was running the Node matrix (20.x, 22.x, 24.x) in parallel against the same Adobe credentials and the same Runtime namespace.
That caused two classes of failures:

• aio-lib-analytics intermittently failed with 429 Too Many Requests
• aio-lib-runtime intermittently failed because parallel jobs deployed and undeployed the same actions in the same namespace
  There was also a bug in the orchestrator: if one upstream repo failed inside runOne(), the process never changed back to the parent directory, so later repos could be cloned inside the failed repo's directory tree.

What This PR Changes

• adds max-parallel: 1 to the workflow matrix so Node versions run sequentially
• adds fail-fast: false so all matrix jobs complete and report their individual status
• wraps process.chdir(name) in runOne() with a try/finally so the orchestrator always returns to the parent directory even when a repo's e2e test fails

Related Issue

https://jira.corp.adobe.com/browse/ACNA-4510

Motivation and Context

How Has This Been Tested?

• npx jest --verbose
• 19/19 tests passing
• 100% coverage## Screenshots (if appropriate):

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• I have signed the Adobe Open Source CLA.
• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• All new and existing tests passed.