Dockerize chat_backend for server-infra deploy (closes #6).

Add uv packaging, env-driven Django settings, OLLAMA_BASE_URL (LAN GPU
host), DatabaseStorage for prompt/document blobs, compose/CI/deploy
workflows mirroring scha, and ASGI gunicorn+UvicornWorker for WebSockets.

scripts/docker-entrypoint.sh

@@ -0,0 +1,53 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+cd /app/llm_be
+
+wait_for_database() {
+  if [[ -z "${DATABASE_URL:-}" && -z "${DB_HOST:-}" ]]; then
+    return 0
+  fi
+
+  echo "Waiting for database..."
+  for _ in $(seq 1 30); do
+    if SKIP_RAG_INIT=1 uv run python - <<'PY'
+import os
+import sys
+
+os.environ.setdefault("DJANGO_SETTINGS_MODULE", "llm_be.settings")
+os.environ["SKIP_RAG_INIT"] = "1"
+
+import django
+from django.db import connections
+from django.db.utils import OperationalError
+
+django.setup()
+
+try:
+    connections["default"].ensure_connection()
+except OperationalError:
+    sys.exit(1)
+PY
+    then
+      echo "Database is ready."
+      return 0
+    fi
+    sleep 2
+  done
+
+  echo "Database did not become ready in time." >&2
+  exit 1
+}
+
+wait_for_database
+
+# ASGI (HTTP + WebSockets). UvicornWorker required for channels.
+export SKIP_RAG_INIT=1
+uv run python manage.py migrate --noinput
+uv run python manage.py collectstatic --noinput
+unset SKIP_RAG_INIT
+
+exec uv run gunicorn llm_be.asgi:application \
+  -k uvicorn.workers.UvicornWorker \
+  --bind "${GUNICORN_BIND:-0.0.0.0:8000}" \
+  --workers "${GUNICORN_WORKERS:-2}"