fix(test runner): do not count statically skipped tests when sharding

[Yury Semikhatsky (yury-s)]

Summary

• Statically skipped tests (test.skip, test.fixme, describe.skip) no longer count toward shard sizes, so shards with many skipped tests are not starved of real work.
• Fully skipped groups follow the preceding group; if all tests are skipped, they are balanced as before.

Fixes #42875

[github-actions]

Test results for "MCP"

1 failed
❌ [chromium] › mcp/http.spec.ts:105 › http transport browser lifecycle (isolated) @mcp-ubuntu-latest-chromium

8711 passed, 1480 skipped

---

Merge workflow run.

[github-actions]

Test results for "tests 1"

8 flaky

⚠️ [installation tests] › playwright-cli-install-should-work.spec.ts:144 › install command should work with HTTPS proxy for HTTP downloads `@package-installations-ubuntu-latest`
⚠️ [installation tests] › playwright-test-package-managers.spec.ts:19 › npm: @playwright/test should work `@package-installations-ubuntu-latest`
⚠️ [chromium-library] › library/beforeunload.spec.ts:130 › should support dismissing the dialog multiple times `@chromium-ubuntu-22.04-arm-node20`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:301 › should report intercepted service worker requests in HAR `@chromium-ubuntu-22.04-node20`
⚠️ [chromium-library] › library/video.spec.ts:762 › screencast › should work with video+trace `@chromium-ubuntu-22.04-node22`
⚠️ [chromium-page] › page/workers.spec.ts:191 › should attribute network activity for worker inside iframe to the iframe `@chromium-ubuntu-22.04-node22`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@firefox-ubuntu-22.04-node20`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe `@firefox-ubuntu-22.04-node20`

52282 passed, 1241 skipped

---

Merge workflow run.

[github-actions]

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

🟢 The failure looks like a pre-existing flake, not caused by this PR

The only failing test, mcp/http.spec.ts:105, is a known intermittent timeout on main and on unrelated PRs. This PR only changes how the test runner builds shards.

Details

This PR changes packages/playwright/src/runner/testGroups.ts (shard grouping for statically skipped tests), plus a docs line and new tests in tests/playwright-test/shard.spec.ts. None of that runs inside the MCP HTTP transport or its browser lifecycle, and the MCP suite doesn't use sharding logic that could starve or drop this test.

Pre-existing flake / infra

• [chromium] › mcp/http.spec.ts:105 › http transport browser lifecycle (isolated) (mcp-ubuntu-latest-chromium): fails with Test timeout of 30000ms exceeded. In the test-results DB, this test's final verdict flips between runs: 743 runs passed and 5 timed out. At least 13 attempts timed out on unrelated SHAs, where a retry sometimes rescued the run. That includes push runs on main (e.g. run 34879545139, run 33102048038) and PRs feat(mcp): add media emulation #42668, fix(server): dispose temporary iframe element handles after hit-target checks #42655, feat(mcp): expose page-registered WebMCP tools #42613, chore(trace-viewer): only offer "Display Aria" when the trace has both snapshot kinds #42435 and feat(tracing): allow configuring the screenshots size #42283, all with the same timeout signature.

_{Triaged by the Playwright bot - agent run}