chore: cleanup more ct remnants

[Dmitry Gozman (dgozman)]

Follow up to #42168.

[github-actions]

Test results for "MCP"

2 failed
❌ [chrome] › mcp/annotate.spec.ts:386 › should cancel browser_annotate when the MCP request is aborted @mcp-windows-latest-chrome
❌ [chrome] › mcp/annotate.spec.ts:417 › should cancel browser_annotate when the MCP client disconnects @mcp-windows-latest-chrome

8087 passed, 1284 skipped

---

Merge workflow run.

[github-actions]

Hi, I'm the Playwright bot and I took a first look at the CI failures here.

🟢 Both failures are pre-existing flakes — this PR is clear

The two mcp/annotate.spec.ts cancellation tests flip verdict across many unrelated SHAs and PRs, and this PR doesn't touch the MCP annotate flow at all — it only cleans up test-runner plugin/dependency-tracking code.

Details

This PR removes leftover component-testing plumbing: the populateDependencies/clearCache/begin/end plugin hooks, the externalDependencies map in compilationCache.ts, and the populateDependenciesOnList test-server option. None of that is exercised by the MCP annotate cancellation flow.

Pre-existing flake / infra

• [chrome] › mcp/annotate.spec.ts:386 › should cancel browser_annotate when the MCP request is aborted — pre-existing flake. Across the test-results DB it failed 8 of 8241 runs, passing the other 8233, on SHAs unrelated to this PR (e.g. PR feat(chromium): roll to r1237 #42072 6f00d1da, PR Revert "chore(deps-dev): bump fast-uri from 3.1.3 to 3.1.4" #41978 bb336448, PR feat(firefox): roll to r1536 #41859 f3070399, plus several push runs). It also spans multiple engines (chrome, firefox, msedge, chromium), which a plugin-cleanup change wouldn't explain.

• [chrome] › mcp/annotate.spec.ts:417 › should cancel browser_annotate when the MCP client disconnects — pre-existing flake. Failed 9 of 8241 runs, passing the other 8232, again on unrelated SHAs (PR Revert "chore(deps-dev): bump fast-uri from 3.1.3 to 3.1.4" #41978 bb336448, plus push runs c3749105, f4aa8802, d0c1b736, dbb805e6, …) across chrome/firefox/msedge/chromium.

Both are cancellation/disconnect tests — inherently timing-sensitive — and neither was rescued on retry in the runs where they flaked, consistent with a genuine intermittent race rather than a code regression.

_{Triaged by the Playwright bot - agent run}