Fix publish workflow test failures by extracting reusable test infrastructure

[Copilot]

Tests pass in python-test.yml but fail in build.yml during publish because the build workflow lacked Chrome/chromedriver/XVFB setup required by Selenium tests.

Changes

• Created .github/workflows/test-runner.yml: Reusable workflow encapsulating browser test infrastructure (Chrome, chromedriver, XVFB). Supports both source and distribution testing via test-distribution flag.

• Refactored .github/workflows/build.yml: Removed inline test validation. Split into build job (creates artifacts) and validate job (delegates to test-runner with test-distribution: true).

• Refactored .github/workflows/python-test.yml: Replaced duplicated test setup with calls to test-runner for Python 3.10 and 3.12. Extracted lint as separate job.

• Security: Added explicit permissions: contents: read blocks per CodeQL recommendations.

Example Usage

# python-test.yml now simply calls the reusable workflow
jobs:
  test-python-310:
    uses: ./.github/workflows/test-runner.yml
    with:
      python-version: "3.10"
    secrets: inherit

# build.yml validates the built distribution
  validate:
    needs: build
    uses: ./.github/workflows/test-runner.yml
    with:
      python-version: ${{ inputs.python-version }}
      test-distribution: true
    secrets: inherit

This eliminates 56 lines of duplicated setup code while fixing the publish workflow.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Fix Publish / Test flow</issue_title>
<issue_description>Test keeps failing on the publish flow, but the tests work on the python-test flow for code reviews</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Fix Publish / Test flow #6

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[BSd3v]

💃