feat: add declarative site parser engine

RAA.md

@@ -1,6 +1,6 @@
 # RAA deployment
 
-Last updated: 2026-08-10.
+Last updated: 2026-08-11.
 
 RAA is not a separate code project anymore. The source of truth for code and
 documentation is:
@@ -96,20 +96,15 @@ All source records, JSON configs, schedules, errors, raw posts, and media state
 remain in the RAA database. The worker is managed by Docker Compose from
 `/opt/site-parser` and should be healthy as `site-parser-worker`.
 
-PopularAirsoft source id `72`:
+PopularAirsoft source id `72` uses `https://popularairsoft.com/` and the full
+v1 config from `site_parser_worker/README.md`. Its RSS is not used for
+discovery because it mixes short video pages with full articles. HTML discovery
+selects the `The Latest News` cards and the worker then opens every `/news/...`
+detail. Verified live 2026-08-11: nine details parsed, zero errors.
 
-- URL: `https://popularairsoft.com/rss.xml`
-- Config: `{"format":"rss","access":"cloudflare","max_items":10,"min_text_length":50}`
-- `access='cloudflare'` deliberately starts the browser/RuCaptcha flow before
-  fetching the RSS.
-- Keep `follow_links` disabled for this feed. Nine current entries have no
-  body, only title/video, and are intentionally filtered before media upload.
-- Cloudflare cookies and browser User-Agent are reused by `media-uploader` for
-  protected images.
-- Admin image previews use authenticated Telegram-backed
-  `/raw/media/{media_id}` URLs.
-- Verified 2026-08-10: source `ok`; raw post `102` is `storage_ready`; four
-  images are `uploaded`. The AI later rejected the content as non-target.
+Cloudflare session state is returned to the main app and is also reused by
+`media-uploader` for protected images. Admin previews remain authenticated
+Telegram-backed `/raw/media/{media_id}` URLs.
 
 ## RAA Categories