[TayoBus]

Problem

A user was hit by a false positive from the bio-link spam detector: their Telegram profile bio contained a link to their own profile (t.me/<their-username>), and has_suspicious_bio_links flagged it as a promo link because the function never knew whose bio it was checking.

Fix

Dynamic per-user self-whitelist in src/bot/handlers/bio_bait.py:

• has_suspicious_bio_links(bio) now accepts an optional username parameter (the bio owner's Telegram username).
• t.me/<own-username> / telegram.me/<own-username> links (any scheme, case-insensitive) are treated as self-references and skipped.
• Bare @<own-username> mentions are excluded from the suspicious-mention count.
• handle_bio_bait_spam passes user.username from the message sender through to the check.
• The static community whitelists (WHITELISTED_TELEGRAM_PATHS, is_url_whitelisted) are untouched — the self-whitelist is dynamic per user and applies only to the bio check.

Edge cases preserved

• Users without a username (username=None): behavior unchanged.
• Bio with own link plus another channel link: still flagged (only self-references are skipped).
• Private invite links (t.me/+hash): always flagged — they can never be a self-reference.
• Single @own-username + promo hint word: no longer flagged (mention count drops to 0).

Verification

• ruff check: clean
• mypy src/bot/handlers/bio_bait.py: clean
• pytest tests/test_bio_bait.py: 112 passed (10 new tests for the self-whitelist)
• Full suite: 1309 passed