build(docker): slim backend image, static frontend image, -docling variant

Backend (arc53/docsgpt): 4.5 GB compressed -> 0.9 GB with both embedding
models and tiktoken baked in.
- torch/transformers gone from the default install (docling extra only).
- Ubuntu 24.04 ships python3.12: no deadsnakes PPA, no software-properties-
  common; every pin is a wheel, so no gcc/g++/rust in the builder.
- COPY --chown and a prefetch that runs as the process user replace the
  trailing chown -R, which duplicated the 600 MB model layer.
- .dockerignore keeps __pycache__, .coverage, local indexes and .env out.
- EXTRAS build arg (INSTALL_DOCLING kept as an alias); the docling variant
  also bakes docling's layout/table/RapidOCR models (DOCLING_ARTIFACTS_PATH)
  and tesseract, and drops only the discovery documents of Google APIs the
  app never builds.
- FLASK_DEBUG env removed (unused); OCI labels added.

Frontend (arc53/docsgpt-fe): 302 MB Vite dev server -> 25 MB static build
behind nginx. VITE_* variables are injected at container start into
/config.js and read through src/env.ts, so the image no longer needs a
rebuild per deployment; docker-compose.yaml keeps hot reload via the dev
target.

Publishing: every release and develop build now pushes a slim tag and a
-docling tag (docling engine + models + tesseract). docker-compose-hub.yaml
takes DOCSGPT_IMAGE_TAG / DOCSGPT_IMAGE_VARIANT; docker-compose-standalone.yaml
runs the stack from pre-built images without a checkout and is attached to
each release. setup.sh selects the -docling variant for OCR instead of
requiring a local build. A new workflow builds the image on PRs that touch
it and runs verify_offline under --network none; lint checks the exported
requirements match uv.lock.

deployment/docker-compose-hub.yaml

@@ -1,8 +1,14 @@
+# Pre-built images from Docker Hub (mirrored at ghcr.io/arc53).
+#   DOCSGPT_IMAGE_TAG      develop (default, follows main) or a release, e.g. 0.20.0
+#   DOCSGPT_IMAGE_VARIANT  empty (default, slim) or -docling: docling parser engine,
+#                          its models, and tesseract baked in (OCR-ready)
+# Set them in ../.env or the shell. deployment/docker-compose-standalone.yaml is
+# the same stack without a git checkout.
 name: docsgpt-oss
 services:
 
   frontend:
-    image: arc53/docsgpt-fe:develop
+    image: arc53/docsgpt-fe:${DOCSGPT_IMAGE_TAG:-develop}
     environment:
       - VITE_API_HOST=http://localhost:7091
       - VITE_API_STREAMING=${VITE_API_STREAMING:-true}
@@ -15,7 +21,7 @@ services:
 
   backend:
     user: root
-    image: arc53/docsgpt:develop
+    image: arc53/docsgpt:${DOCSGPT_IMAGE_TAG:-develop}${DOCSGPT_IMAGE_VARIANT:-}
     env_file:
       - ../.env
     environment:
@@ -38,7 +44,7 @@ services:
 
   worker:
     user: root
-    image: arc53/docsgpt:develop
+    image: arc53/docsgpt:${DOCSGPT_IMAGE_TAG:-develop}${DOCSGPT_IMAGE_VARIANT:-}
     # `parsing` queue carries read_document/parse_document; required for its await to resolve.
     command: celery -A application.app.celery worker -l INFO -B -Q docsgpt,parsing,embeddings
     env_file: