feat(b3): ml-worker becomes optional — embed-only role, decoupled GPU coordination, cpu-embed switch

The ml-worker's ONLY processing role is now the CPU whole-image embed fallback
(tag_and_embed renamed embed_image — Camie tagging was retired #1189 and the
name kept implying otherwise; videos were already handled agent-style: frame
sampling + mean-pool). Detection/cropping/CCIP stay GPU-agent-only, and their
completion is judged per-pipeline: ccip by gpu_job rows, siglip by concept
regions at the current model version — never by image_record.siglip_embedding.
A CPU embed therefore can NEVER close crop work for the agent (regression test
pins this; only the whole-image 'embed' job, the same artifact, is satisfied).

Making removal actually safe (operator will drop the container):
- GPU-queue coordination (enqueue_gpu_backfill, recover_orphaned_gpu_jobs,
  reprocess_gpu_jobs) moved verbatim to tasks/gpu_queue.py on the maintenance
  quick lane — it lived on the 'ml' queue only by module colocation, which made
  the ml-worker a hard dependency of the whole agent pipeline.
- New ml_settings.cpu_embed_enabled (migration 0074, default ON so agent-less
  installs keep working): OFF stops the four import hooks queueing embed work
  nothing will consume and no-ops the manual backfill; switch lives on the
  renamed 'CPU embedding backfill' card.
- NB heads training / auto-apply still run on the ml image (sklearn) — a stack
  that removes the container gives those up too.

Deploy note: in-flight messages under the old task names are dropped by the
new workers; the 60s orphan sweep + hourly backfill re-fire under the new
names immediately.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM

backend/app/tasks/import_file.py

@@ -228,15 +228,17 @@ def _do_import(session, task, import_task_id: int) -> dict:
     # Enqueue thumbnail + ML for newly imported AND superseded images
     # (a superseded row has cleared ML + no thumbnail).
     if result.status in ("imported", "superseded"):
-        from .ml import tag_and_embed
+        from .ml import cpu_embed_enabled, embed_image
         from .thumbnail import generate_thumbnail
 
+        do_embed = cpu_embed_enabled()
         ids = list(result.member_image_ids)
         if result.image_id is not None and result.image_id not in ids:
             ids.append(result.image_id)
         for img_id in ids:
             generate_thumbnail.delay(img_id)
-            tag_and_embed.delay(img_id)
+            if do_embed:
+                embed_image.delay(img_id)
 
     # If this was the last task in the batch, mark the batch complete.
     remaining = session.execute(