Phase 6: sync jobs, inquiry and order forms

- Sync scheduler in the content service with a per-job cache (sync_cache),
  sync_runs reporting and Healthchecks pings. Jobs: Archidekt, BoardGameGeek,
  cover art (Open Library, TMDB, AniList), Suwayomi and Kavita reading
  progress, Bambuddy filament and Homebox material stock. A note's own value
  always wins over a synced one.
- PersonalReader: updates only Library notes with a matching anilistId,
  changes to Greg's notes are held for review; Reading series with no note
  show as suggestions on the admin Overview.
- /api/inquiries and /api/orders: Turnstile, rate limits, a saved copy in the
  content service (intake token) plus email over SMTP. Orders are re-checked
  against the live notes: accepted only when the category's ordersOpen is
  true and the line has capacity.
- Admin Inbox for saved submissions. Options' inStock is optional so synced
  stock can fill it. All settings from process.env; .env.example per service.
- Sources could not be reached from the build environment; parsers are tested
  against fixtures labelled as not live.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012aSbo2D1qP8cMFQCykf7kx

services/content/test/sync/fixtures/README.md

@@ -0,0 +1,22 @@
+# Sync test fixtures
+
+Small sample payloads written by hand from each API's documented response
+shape, for the parsing tests. They are **not** live responses: no request was
+made to Archidekt, BoardGameGeek, Open Library, TMDB, AniList, Suwayomi,
+Kavita, Bambuddy or Homebox to make them,
+and ids, names and image addresses are illustrative.
+
+| File | Shape of |
+| --- | --- |
+| `archidekt-deck.sample.json` | `GET https://archidekt.com/api/decks/<id>/` |
+| `bgg-thing.sample.xml` | `GET https://boardgamegeek.com/xmlapi2/thing?id=…` |
+| `openlibrary-search.sample.json` | `GET https://openlibrary.org/search.json?title=…` |
+| `tmdb-search.sample.json` | `GET https://api.themoviedb.org/3/search/movie?query=…` |
+| `anilist-media.sample.json` | `POST https://graphql.anilist.co` with the `Media` query |
+| `suwayomi-library.sample.json` | `POST <SUWAYOMI_URL>/api/graphql` with `LIBRARY_QUERY` (Suwayomi-Server's GraphQL schema) |
+| `anilist-page.sample.json` | `POST https://graphql.anilist.co` with the batched `Page { media(id_in:) }` query |
+| `kavita-series.sample.json` | `POST <KAVITA_URL>/api/Series/all-v2` (SeriesDto list) |
+| `kavita-metadata.sample.json` | `GET <KAVITA_URL>/api/Series/metadata?seriesId=`, one answer per series id |
+| `bambuddy-spools.sample.json` | **A guess.** Bambuddy's API is not documented anywhere this project has; this only exercises `mapBambuddySpools`' tolerant reading. |
+| `homebox-labels.sample.json` | `GET <HOMEBOX_URL>/api/v1/labels` |
+| `homebox-items.sample.json` | `GET <HOMEBOX_URL>/api/v1/items?labels=…` |