feat(mcp): default codegen language from PW_LANG_NAME

[Simon Knott (Skn0tt)]

Language ports already set PW_LANG_NAME when they spawn the driver (python / java / csharp). MCP codegen now defaults from that, so playwright mcp under playwright-python emits Python without needing --codegen=python.

Also wires PLAYWRIGHT_MCP_CODEGEN through configFromEnv, which the flag docs already advertised.

[github-actions]

Test results for "MCP"

1 failed
❌ [firefox] › mcp/cli-core.spec.ts:140 › uncheck @mcp-windows-latest-firefox

8062 passed, 1284 skipped

---

Merge workflow run.

[github-actions]

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

🟢 The one failure is a pre-existing flake — this PR looks clear

[firefox] › mcp/cli-core.spec.ts:140 › uncheck (mcp-windows-latest-firefox) is unrelated to this change. The PR only adjusts the default codegen language (config.ts, program.ts), which the uncheck CLI flow never touches, and the test has a track record of flaking on SHAs this PR can't be responsible for.

Details

Overall: one failed test in the merged report, no real failures traced back to the diff. This PR defaults MCP codegen from PW_LANG_NAME and wires PLAYWRIGHT_MCP_CODEGEN through configFromEnv — it doesn't alter check/uncheck behavior. On CI PW_LANG_NAME is unset, so the default stays typescript and behavior is unchanged.

Pre-existing flake / infra

• [firefox] › mcp/cli-core.spec.ts:140 › uncheck — pre-existing flake. In the test-results DB this test's final verdict flips across runs on firefox: failed 2 of 1722 runs, passed the other 1720, including on unrelated work — PR test(websocket): cover setExtraHTTPHeaders and locale on the handshake request #42037 (sha dc298405) and a July 30 main push (sha 30d5b2ec). Scoped to the mcp-windows-latest-firefox bot it's 2 red in 413 runs. It also flaked once on msedge (1/460). None of those SHAs involve this PR, and the diff doesn't reach the CLI uncheck path.

_{Triaged by the Playwright bot - agent run}