docs(decisions): add DR-001 for unit test design

[clanghans]

Summary

• Adds docs/decisions/ directory with an index and the first Decision Record
• DR-001-Infra documents four design decisions from the unit test infrastructure work in PR feat(test): add unit test infrastructure with py_itf_unittest #94:
  1. Dedicated py_itf_unittest macro (thin wrapper, no plugin machinery)
  2. Surgical Bazel target splitting for atomic dependency scoping
  3. Shared main.py bootstrap pattern for pytest under Bazel
  4. pytest-mock over unittest.mock
• Naming convention and index structure follow the Eclipse S-CORE DR convention (DR-{number}-{category}.md, glob-based toctree per category)
• Wires the new decisions/index into the Sphinx toctree

Related: #94
Discussion: https://github.com/orgs/eclipse-score/discussions/2867

Test plan

• Sphinx build succeeds (bazel build //docs)
• DR renders correctly in the generated docs site
• Links to PR feat(test): add unit test infrastructure with py_itf_unittest #94 and the GitHub discussion are reachable

[Copilot]

Pull request overview

Adds a new “Decisions” documentation section to capture design decision records (DRs) for ITF, starting with DR-001 describing the unit test infrastructure decisions introduced in PR #94.

Changes:

• Add docs/decisions/ with an index page and the first decision record (DR-001-infra).
• Wire the new decisions section into the main Sphinx docs toctree.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
docs/index.rst	Adds decisions/index to the site-level toctree so the section appears in the docs build.
docs/decisions/index.rst	Introduces the Decisions landing page and a globbed toctree for infra DRs.
docs/decisions/DR-001-infra.md	Documents the rationale and selected options for the unit test infrastructure design from PR #94.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[pahmann]

Thanks for the DR. This looks like a solid and pragmatic approach.

I’m not a Bazel nor py_test expert, so please take this more as a question from an OSS perspective: have we considered leveraging existing pytest integrations in the Bazel ecosystem (e.g. py_pytest_test-style rules) instead of maintaining a custom main.py bootstrap?

The current approach works well, but introducing a custom macro + bootstrap adds project-specific infrastructure. Aligning more closely with community-maintained rules could help reduce maintenance overhead and lower the entry barrier for external contributors who are already familiar with standard pytest setups.

This wouldn’t necessarily require changing the external model. Maybe keep py_itf_unittest as a thin wrapper, but delegate the implementation to an existing pytest-aware rule.

Not a blocker, just a suggestion to consider for better long-term alignment with the Bazel OSS ecosystem.