ROX-32888: Use ACS bot token to create PR

[mtodor]

Description

Uses RHACS_BOT_GITHUB_TOKEN instead of the default GITHUB_TOKEN in the peter-evans/create-pull-request step of the model-evaluation workflow. PRs created with the default token do not trigger pull_request workflows (GitHub's design to prevent infinite loops), which meant required CI checks (build, test, e2e, style, smoke, wiremock-test) were not running on auto-generated evaluation PRs.

Also changes the weekly schedule from Monday to Tuesday at 6:00 UTC - to test trigger.

Validation

• actionlint passes on the modified workflow file
• Verify CI checks trigger on the next auto-generated evaluation PR (after merge)

AI-assisted development prompts

This PR was developed with AI assistance. Below are the prompts used:

Prompt: "Weekly job is running now, but we have a problem that created PR is not running tests. Here is an example: #114 - if you investigate that PR, you will see that required tests are not triggered. What should be done to fix that?"
Response: Identified the root cause: peter-evans/create-pull-request using the default GITHUB_TOKEN does not trigger pull_request workflows. Recommended using a PAT or GitHub App token instead.

Prompt: "Ok. I have token in secret: RHACS_BOT_GITHUB_TOKEN - use that"
Response: Added token: ${{ secrets.RHACS_BOT_GITHUB_TOKEN }} to the Create Pull Request step.

Prompt: "Let's switch to Tuesday - so that we can test it next day."
Response: Changed cron schedule from 0 6 * * 1 (Monday) to 0 6 * * 2 (Tuesday).

[codecov-commenter]

❌ 2 Tests Failed:

Tests completed	Failed	Passed	Skipped
361	2	359	12

View the full list of 2 ❄️ flaky test(s)

::policy 1

Flake rate in main: 100.00% (Passed 0 times, Failed 16 times)

Stack Traces | 0s run time

- test violation 1
- test violation 2
- test violation 3

::policy 4

Flake rate in main: 100.00% (Passed 0 times, Failed 16 times)

Stack Traces | 0s run time

- testing multiple alert violation messages 1
- testing multiple alert violation messages 2
- testing multiple alert violation messages 3

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}

[github-actions]

E2E Test Results

Commit: fa3d295
Workflow Run: View Details
Artifacts: Download test results & logs

=== Evaluation Summary ===

  ✓ list-clusters (assertions: 3/3)
  ✓ cve-detected-workloads (assertions: 3/3)
  ✓ cve-detected-clusters (assertions: 3/3)
  ~ cve-nonexistent (assertions: 2/3)
      - MaxToolCalls: Too many tool calls: expected <= 5, got 7
  ✓ cve-cluster-does-exist (assertions: 3/3)
  ~ cve-cluster-does-not-exist (assertions: 2/3)
      - ToolsUsed: Required tool not called: server=stackrox-mcp, tool=, pattern=list_clusters
  ✗ cve-clusters-general (assertions: 3/3)
      one or more verification steps failed
  ✓ cve-cluster-list (assertions: 3/3)
  ✓ cve-log4shell (assertions: 3/3)
  ✓ cve-multiple (assertions: 3/3)
  ✓ rhsa-not-supported (assertions: 2/2)

Tasks:      10/11 passed (90.91%)
Assertions: 30/32 passed (93.75%)
Tokens:     ~58232 (estimate - excludes system prompt & cache)
MCP schemas: ~12738 (included in token total)
Agent used tokens:
  Input:  14453 tokens
  Output: 22787 tokens
Judge used tokens:
  Input:  62504 tokens
  Output: 68610 tokens