Add Postgres store, path safety, and the crawl indexer

Store: crawl generations as the sync cursor. Diffs compare generations
on entity identity plus a content digest, never on upstream timestamps —
GET /course-rooms/{id}/board returns request time as updatedAt for most
elements, so a timestamp cursor would report every board as changed on
every crawl. Identity diffing also yields deletions, which no timestamp
scheme can. A per-course crawl carries the other courses' rows forward
so every completed generation is a complete picture and any two diff
directly; without that a partial crawl reads as a mass deletion.

FTS uses the german dictionary with weighted title/body, plus a pg_trgm
arm because stemming will not match "Datenschutz" inside
"Datenschutzgrundverordnung" and German compounds make that the common
case. file_texts is keyed by file record id and deliberately outlives
generations: records are immutable upstream, so text extracted once is
valid forever and a re-crawl of unchanged content costs nothing.

Store.open returns undefined instead of throwing when Postgres is
unreachable — the index is an accelerator, and a Pi that loses its
database should get slower, not broken.

core/paths.ts is the security boundary for the mirror. Course titles,
card titles and filenames are all user-supplied upstream, so this is
where a hostile name stops being text and becomes a path. Two bugs found
by its own tests: "///" produced "---" instead of falling back, and dot
runs survived mid-component. Now no ".." can survive anywhere, which
makes the invariant checkable rather than a claim about ordering.

Indexer coalesces concurrent refreshes onto one run and enforces a
minimum interval, since a full crawl is ~270 requests from an account
that looks like a student.

9 store tests against a real Postgres (mocks would test nothing here)
and 13 path tests; 47 total.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

docker-compose.yml

@@ -6,6 +6,27 @@
 # `networks` block at the bottom and deploy/Caddyfile.snippet.
 
 services:
+  # Dev/standalone Postgres. On the Pi, point DATABASE_URL at the existing
+  # instance instead and remove this service — the schema lives in its own
+  # database and user, so it coexists with whatever else is already there.
+  postgres:
+    image: postgres:17-alpine
+    container_name: schulcloud-mcp-db
+    restart: unless-stopped
+    environment:
+      POSTGRES_DB: schulcloud
+      POSTGRES_USER: schulcloud
+      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-schulcloud}
+    volumes:
+      - pgdata:/var/lib/postgresql/data
+    networks:
+      - caddy
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U schulcloud -d schulcloud"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+
   schulcloud-mcp:
     build: .
     image: schulcloud-mcp:latest
@@ -15,6 +36,14 @@ services:
     environment:
       PORT: 8080
       BIND_HOST: 0.0.0.0
+      MIRROR_DIR: /data/mirror
+    depends_on:
+      postgres:
+        condition: service_healthy
+    volumes:
+      # The mirror is the one thing this server writes; everything else stays
+      # read-only, so it gets its own volume rather than loosening read_only.
+      - mirror:/data/mirror
     # No ports are published to the host: Caddy reaches the container over the
     # shared Docker network, so the only way in from the internet is through
     # Caddy's TLS and this server's bearer check.
@@ -35,6 +64,10 @@ services:
     cap_drop:
       - ALL
 
+volumes:
+  pgdata:
+  mirror:
+
 networks:
   caddy:
     # Set to true once this joins the network your existing Caddy already uses,