Add LinkedIn ingestion connector (lazy Playwright)

[devin-ai-integration]

Summary

Completes the ingestion connector set from the production-ready decomposition — the HTTP/stdlib connectors landed in #11; LinkedIn was deferred because it needs Playwright. Keeps the established pattern: Playwright is imported lazily (inside _get_browser + under TYPE_CHECKING), so it stays an optional dependency and adds no new hard dep to main.

• ingestion/linkedin.py — LinkedInConnector(BaseConnector) (source_name = "linkedin"): headcount + hiring scrape via Playwright, with _parse_headcount turning size strings into {min, max, exact}. Credentials fall back to settings; fetch() degrades gracefully ({"found": False, ...}) when creds are missing or scraping errors.
• config.py — adds linkedin_email / linkedin_password (both None by default), matching the existing connector-credential convention.

Following #11's convention, the connector is imported by module path rather than re-exported from ingestion/__init__.py.

Testing

• tests/test_ingestion/test_linkedin.py — connector identity/BaseConnector contract, credential handling, graceful degradation without creds, and a parametrized _parse_headcount table ("51-200" → min/max, "10,000+" → min, "42" → exact, empty/garbage → all None). No browser launched, no network.
• ruff check . clean; full suite 73 passed.

Link to Devin session: https://app.devin.ai/sessions/7eecbb4c058a4651a15001ade15f0387
Requested by: @askmy-stack

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring