feat(search): zero-quota InnerTube scrape source + search_source toggle

Live YouTube search can now use YouTube's internal InnerTube endpoint instead
of search.list, materialising results through the same enrich/provenance path
at zero API quota (search.list costs 100 units/page; scrape costs nothing, only
the cheap shared videos.list enrich is charged).

- youtube/search_scrape.py: InnerTube search returning the same page shape as
  YouTubeClient.search_videos (items + continuation cursor); SOCS consent cookie,
  videoRenderer walk (channels/playlists/Shorts shelves naturally excluded),
  type:video filter, cached InnerTube key/version with constant fallbacks.
- routes/search.py: admin-selectable source (search_source, default scrape).
  Scrape path skips the 100-unit budget pre-check and logs a zero-cost search
  event so the per-user daily cap still counts it; api path unchanged. Response
  carries the active source.
- sysconfig/config: new search_source key (scrape|api).
- quota.log_action: record a zero-cost action event for per-user rate limits.

backend/app/routes/search.py

@@ -1,10 +1,12 @@
 """Live YouTube search.
 
-Searches YouTube directly (search.list) and materialises the results into the shared catalog
-so they render with the normal feed cards + in-app player and gain per-user state (watch /
-save / resume). search.list costs **100 quota units per call**, so it's the most expensive
-read by far: it's gated behind a per-user daily cap + a budget pre-check, and only real users
-(never the shared demo account) may call it.
+Searches YouTube directly and materialises the results into the shared catalog so they render
+with the normal feed cards + in-app player and gain per-user state (watch / save / resume). The
+source is admin-selectable (``search_source``): **scrape** (InnerTube, the website's own
+endpoint, zero API quota — the default) or **api** (search.list, **100 quota units per page**).
+Either way it's gated behind a per-user daily cap, the API source additionally behind a budget
+pre-check, and only real users (never the shared demo account) may call it. The result ids are
+enriched through the cheap shared videos.list path (1 unit / 50) regardless of source.
 
 Ingested results are flagged (videos.via_search / channels.from_search) so the Library can hide
 this search-discovered "noise" by default. Shorts and currently-live/upcoming videos are never
@@ -24,6 +26,7 @@ from app.models import Channel, User, Video, VideoState
 from app.routes.feed import _serialize, feed_columns
 from app.sync.subscriptions import apply_channel_details
 from app.sync.videos import _insert_stubs, apply_video_details, parse_dt
+from app.youtube import search_scrape
 from app.youtube.client import YouTubeClient, YouTubeError
 from app.youtube.shorts import make_client, probe_is_short
 
@@ -100,7 +103,13 @@ def search_youtube(
     if not term:
         raise HTTPException(status_code=400, detail="A search term is required.")
 
-    # Guard the shared quota BEFORE spending: per-user daily cap, then today's budget.
+    # Source: "scrape" (InnerTube, no API quota) or "api" (search.list, 100 units/page).
+    source = (sysconfig.get_str(db, "search_source") or "scrape").lower()
+    if source != "scrape":
+        source = "api"
+
+    # Per-user daily cap applies to both sources (anti-abuse); with scrape it's the only guard
+    # since the search itself spends no quota.
     daily_limit = sysconfig.get_int(db, "search_daily_limit_per_user")
     used = quota.actions_today(db, user.id, quota.QuotaAction.VIDEOS_SEARCH)
     if used >= daily_limit:
@@ -108,7 +117,8 @@ def search_youtube(
             status_code=429,
             detail=f"You've reached today's YouTube search limit ({daily_limit}). Try again tomorrow.",
         )
-    if not quota.can_spend(db, SEARCH_COST):
+    # The 100-unit budget pre-check only applies to the API source.
+    if source == "api" and not quota.can_spend(db, SEARCH_COST):
         raise HTTPException(
             status_code=429,
             detail="The shared YouTube quota for today is used up. Try again tomorrow.",
@@ -116,9 +126,14 @@ def search_youtube(
 
     with YouTubeClient(db, user) as yt:
         try:
-            with quota.attribute(user.id, quota.QuotaAction.VIDEOS_SEARCH):
-                page = yt.search_videos(term, page_token=cursor)
-        except YouTubeError as exc:
+            if source == "api":
+                with quota.attribute(user.id, quota.QuotaAction.VIDEOS_SEARCH):
+                    page = yt.search_videos(term, page_token=cursor)
+            else:
+                page = search_scrape.search(term, continuation=cursor)
+                # Zero-cost, but logged so the per-user daily cap above keeps counting it.
+                quota.log_action(db, user.id, quota.QuotaAction.VIDEOS_SEARCH)
+        except (YouTubeError, search_scrape.ScrapeError) as exc:
             raise HTTPException(status_code=502, detail=f"YouTube search failed: {exc}")
 
         # Drop currently-live/upcoming results (search can't filter them and we never ingest
@@ -134,6 +149,7 @@ def search_youtube(
                 "items": [],
                 "next_cursor": page["next_page_token"],
                 "has_more": bool(page["next_page_token"]),
+                "source": source,
             }
 
         # 1) Channels are the FK target — create stubs for unknown ones (flagged from_search),
@@ -201,4 +217,5 @@ def search_youtube(
         "items": _serialize_ids(db, user, ordered),
         "next_cursor": page["next_page_token"],
         "has_more": bool(page["next_page_token"]),
+        "source": source,
     }