Decouple per-service Alembic migrations and fix cross-DB queries

Each service (blog, market, cart, events, federation, account) now owns
its own database schema with independent Alembic migrations. Removes the
monolithic shared/alembic/ that ran all migrations against a single DB.

- Add per-service alembic.ini, env.py, and 0001_initial.py migrations
- Add shared/db/alembic_env.py helper with table-name filtering
- Fix cross-DB FK in blog/models/snippet.py (users lives in db_account)
- Fix cart_impl.py cross-DB queries: fetch products and market_places
  via internal data endpoints instead of direct SQL joins
- Fix blog ghost_sync to fetch page_configs from cart via data endpoint
- Add products-by-ids and page-config-ensure data endpoints
- Update all entrypoint.sh to create own DB and run own migrations
- Cart now uses db_cart instead of db_market
- Add docker-compose.dev.yml, dev.sh for local development
- CI deploys both rose-ash swarm stack and rose-ash-dev compose stack
- Fix Quart namespace package crash (root_path in factory.py)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

CLAUDE.md

@@ -0,0 +1,74 @@
+# Art DAG Monorepo
+
+Federated content-addressed DAG execution engine for distributed media processing with ActivityPub ownership and provenance tracking.
+
+## Project Structure
+
+```
+core/       # DAG engine (artdag package) - nodes, effects, analysis, planning
+l1/         # L1 Celery rendering server (FastAPI + Celery + Redis + PostgreSQL)
+l2/         # L2 ActivityPub registry (FastAPI + PostgreSQL)
+common/     # Shared templates, middleware, models (artdag_common package)
+client/     # CLI client
+test/       # Integration & e2e tests
+```
+
+## Tech Stack
+
+Python 3.11+, FastAPI, Celery, Redis, PostgreSQL (asyncpg for L1), SQLAlchemy, Pydantic, JAX (CPU/GPU), IPFS/Kubo, Docker Swarm, HTMX + Jinja2 for web UI.
+
+## Key Commands
+
+### Testing
+```bash
+cd l1 && pytest tests/                    # L1 unit tests
+cd core && pytest tests/                  # Core unit tests
+cd test && python run.py                  # Full integration pipeline
+```
+- pytest uses `asyncio_mode = "auto"` for async tests
+- Test files: `test_*.py`, fixtures in `conftest.py`
+
+### Linting & Type Checking (L1)
+```bash
+cd l1 && ruff check .                     # Lint (E, F, I, UP rules)
+cd l1 && mypy app/types.py app/routers/recipes.py tests/
+```
+- Line length: 100 chars (E501 ignored)
+- Mypy: strict on `app/types.py`, `app/routers/recipes.py`, `tests/`; gradual elsewhere
+- Mypy ignores imports for: celery, redis, artdag, artdag_common, ipfs_client
+
+### Docker
+```bash
+docker build -f l1/Dockerfile -t celery-l1-server:latest .
+docker build -f l1/Dockerfile.gpu -t celery-l1-gpu:latest .
+docker build -f l2/Dockerfile -t l2-server:latest .
+./deploy.sh                               # Build, push, deploy stacks
+```
+
+## Architecture Patterns
+
+- **3-Phase Execution**: Analyze -> Plan -> Execute (tasks in `l1/tasks/`)
+- **Content-Addressed**: All data identified by SHA3-256 hashes or IPFS CIDs
+- **Services Pattern**: Business logic in `app/services/`, API endpoints in `app/routers/`
+- **Types Module**: Pydantic models and TypedDicts in `app/types.py`
+- **Celery Tasks**: In `l1/tasks/`, decorated with `@app.task`
+- **S-Expression Effects**: Composable effect language in `l1/sexp_effects/`
+- **Storage**: Local filesystem, S3, or IPFS backends (`storage_providers.py`)
+
+## Auth
+
+- L1 <-> L2: scoped JWT tokens (no shared secrets)
+- L2: password + OAuth SSO, token revocation in Redis (30-day expiry)
+- Federation: ActivityPub RSA signatures (`core/artdag/activitypub/`)
+
+## Key Config Files
+
+- `l1/pyproject.toml` - mypy, pytest, ruff config for L1
+- `l1/celery_app.py` - Celery initialization
+- `l1/database.py` / `l2/db.py` - SQLAlchemy models
+- `l1/docker-compose.yml` / `l2/docker-compose.yml` - Swarm stacks
+
+## Tools
+
+- Use Context7 MCP for up-to-date library documentation
+- Playwright MCP is available for browser automation/testing