fix(platform): detect installed GitHub Copilot clients

[tirth8205]

What changed

Extract only the validated GitHub Copilot detection fix from draft #658:

• Require a real Copilot extension instead of treating any ~/.vscode directory as Copilot.
• Detect user-installed and bundled Copilot/Copilot Chat extensions.
• Cover released VS Code application layouts and executable-relative extension paths.

No Google Antigravity behavior from #658 is included.

Attribution and client evidence

The commit credits npkriami18 <pranaynannuri@gmail.com> as co-author.

The original #658 validation used released VS Code 1.127 with bundled Copilot Chat 0.55: detection succeeded and the generated workspace MCP configuration loaded without a parse error. Copilot CLI discovery, invocation, idempotent reinstall, and surgical uninstall were also validated there. This PR isolates only the detection portion that applies to current main.

Validation

• Fresh local platform/install/uninstall suite: 241 passed.
• Ruff on changed implementation/tests: passed.
• git diff --check: passed.

This PR must pass its complete hosted Python, Windows, lint, type, security, schema, review, CodeQL, and GitGuardian matrix before merge.

Extracted from #658

[github-actions]

code-review-graph review

Overall risk: 0.35 (LOW) — 6 changed function(s)/class(es), 0 affected flow(s), 2 test gap(s)

Risk-scored changes

Risk	Level	Symbol	Location	Tested
0.35	low	code_review_graph/skills.py::_copilot_vscode_detected	code_review_graph/skills.py:35	yes
0.30	low	tests/test_skills.py::TestCopilotPlatform._which	tests/test_skills.py:1702	no
0.25	low	tests/test_skills.py::TestCopilotPlatform.test_copilot_detects_vscode_bundled_extension	tests/test_skills.py:1688	(test)
0.05	low	tests/test_skills.py::TestCopilotPlatform	tests/test_skills.py:1581	no
0.05	low	tests/test_skills.py::TestCopilotPlatform.test_copilot_included_in_all_when_detected	tests/test_skills.py:1672	(test)
0.05	low	tests/test_skills.py::TestCopilotPlatform.test_copilot_not_detected_from_vscode_alone	tests/test_skills.py:1711	(test)

Test gaps

• tests/test_skills.py::TestCopilotPlatform (tests/test_skills.py:1581)
• tests/test_skills.py::TestCopilotPlatform._which (tests/test_skills.py:1702)

Token savings: this graph-backed report used ~61,312 fewer tokens (~98%) than reading every changed file in full (estimated, chars/4 approximation).

---

Powered by code-review-graph — local-first analysis; no code leaves the CI runner.