feat: support timeouts in the send_request context helper (#2138)

[aryansk]

Closes #2138

Summary

SendRequestFunction (used via context.send_request() in request handlers) only accepted url, method, payload and headers, so a handler could not bound how long an extra HTTP call may take. All four HTTP client implementations (HttpxHttpClient, CurlImpersonateHttpClient, ImpitHttpClient, PlaywrightHttpClient) already support a per-request timeout: timedelta | None — this PR just exposes it through the public contract.

Changes

• src/crawlee/_types.py: add timeout: timedelta | None = None to the SendRequestFunction protocol + docstring.
• src/crawlee/crawlers/_basic/_basic_crawler.py: thread timeout through the _prepare_send_request_function closure into HttpClient.send_request.
• tests/unit/crawlers/_basic/test_basic_crawler.py: regression test — context.send_request('/slow?delay=2', timeout=timedelta(milliseconds=100)) raises asyncio.TimeoutError.

Validation

• pytest tests/unit/crawlers/_basic → 112 passed, 5 skipped (3.10/3.11-only features)
• ruff check + ruff format --check clean on all touched files

Note: no new dependency; timeout uses the existing timedelta convention across the Python HTTP clients (the JS implementation names it timeoutMillis, but the Python clients are timedelta-based).

[Battleplus]

Hi! I've opened a PR that supersedes this one for #2138: #2174.

It implements the same feature (exposing timeout through SendRequestFunction / context.send_request()), and additionally fixes a bug your draft would hit: PlaywrightHttpClient.send_request passes timeout.total_seconds() to Playwright's APIRequestContext.fetch, which expects milliseconds — so any timeout passed on a Playwright crawler would be ~1000x too short. The new PR also adds a regression test for that conversion plus the real end-to-end timeout test.

Feel free to close this draft in favor of #2174, or let me know if you'd prefer to take the changes over. Thanks for the original implementation!