Add TweetClaw export source for offline sentiment runs

[kriptoburak]

Summary

• add a small TweetClaw export loader for JSON, JSONL, NDJSON, and CSV files
• let scrape_all include saved TweetClaw posts as a TweetClaw platform source
• document the offline export workflow for reviewed X/Twitter source posts
• add unittest coverage for JSONL, wrapped JSON, and CSV exports

Validation

• python3 -m py_compile app/tweetclaw_import.py app/scraper.py tests/test_tweetclaw_import.py
• python3 -m unittest discover tests
• python3 -m black --check --line-length 100 app/tweetclaw_import.py tests/test_tweetclaw_import.py
• python3 smoke check for extract_tweetclaw_texts
• git diff --check

Notes:

• python3 -m black --check --line-length 100 app/tweetclaw_import.py app/scraper.py tests/test_tweetclaw_import.py would reformat pre-existing app/scraper.py style, so I limited the Black check to the new files.
• npx --yes markdown-link-check README.md --quiet reports pre-existing README links for LICENSE and localhost; the new TweetClaw link is reachable.