The previous parser treated any non-empty non-header line as a bullet,
so 'plain paragraph' answers were silently rewrapped as <ul><li>. Now
we only count a line if it had a recognized bullet marker (- * • ·);
if none did, fall back to raw + <pre> as originally intended.
Tests cover dash/asterisk/unicode bullets, blank-line tolerance,
Summary:/TL;DR header stripping, HTML escaping, single-bullet output,
the no-marker passthrough path, and the IsQuestionReaction set.
Now triggers on ❓❔ ⁉ ⁉️🤔 ? ？ — covers the obvious red/white/thinking
variants, the exclamation-question combo (with and without VS16), and
plain ascii / fullwidth question marks for keyboard users.
When a user reacts ❓ on one of Pete's posts, fetch the article body,
ask Ollama for a 3-bullet summary, and post it as a threaded reply
rooted at the original story event. Per-process cooldown of 5min per
story keeps repeated reactions from re-summarizing.
- ingestion.FetchArticleBody: visible <p> text capped at 8000 chars
- classifier.OllamaClient.GenerateText: non-JSON variant
- storage.GetStoryByGUID: full row lookup
- matrix.PostThreadedReply: m.thread + m.in_reply_to fallback
- poster.SetReactionCallback: optional hook fired after recording
- New package: internal/explainer
