Refactor monolithic modules and harden booking, email, and auth infrastructure.

Split oversized frontend API client, email service, and admin/booking pages into focused modules while preserving import surfaces, and add Redis-backed queues, stale booking cleanup, stronger auth, and scale deployment configs.

Co-authored-by: Cursor <cursoragent@cursor.com>

deploy/docker-compose.scale.yml

@@ -0,0 +1,81 @@
+# Example docker-compose for running the Spanglish API as multiple replicas
+# behind nginx, with Redis for shared state and Postgres as the database.
+#
+# This is a starting point, not a turnkey production setup. It expects a
+# Dockerfile at backend/Dockerfile that builds the API and runs it on PORT.
+#
+# Bring it up with N API replicas:
+#   docker compose -f deploy/docker-compose.scale.yml up --build --scale api=3
+#
+# No em dashes are used in this file by design.
+
+services:
+  postgres:
+    image: postgres:16-alpine
+    environment:
+      POSTGRES_USER: spanglish
+      POSTGRES_PASSWORD: spanglish
+      POSTGRES_DB: spanglish
+    # Raise max_connections if DB_POOL_MAX * replicas approaches the default 100.
+    command: ["postgres", "-c", "max_connections=200"]
+    volumes:
+      - pgdata:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U spanglish"]
+      interval: 5s
+      timeout: 3s
+      retries: 10
+
+  redis:
+    image: redis:7-alpine
+    command: ["redis-server", "--appendonly", "yes"]
+    volumes:
+      - redisdata:/data
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 5s
+      timeout: 3s
+      retries: 10
+
+  api:
+    build:
+      context: ../backend
+    environment:
+      NODE_ENV: production
+      PORT: "3001"
+      DB_TYPE: postgres
+      DATABASE_URL: postgresql://spanglish:spanglish@postgres:5432/spanglish
+      DB_POOL_MAX: "15"
+      REDIS_URL: redis://redis:6379
+      JWT_SECRET: change-me-to-a-strong-secret
+      FRONTEND_URL: http://localhost:8080
+      # Optional S3-compatible storage so uploads are shared across replicas.
+      # If you omit these, mount a shared volume at /app/uploads on every replica.
+      # S3_ENDPOINT: http://garage:3900
+      # S3_REGION: garage
+      # S3_BUCKET: spanglish-media
+      # S3_ACCESS_KEY_ID: ""
+      # S3_SECRET_ACCESS_KEY: ""
+      # S3_PUBLIC_URL: http://localhost:8080/media
+    expose:
+      - "3001"
+    depends_on:
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+  # Load balancer across the scaled api replicas. nginx resolves the "api"
+  # service name via Docker's embedded DNS, which round-robins across replicas.
+  lb:
+    image: nginx:alpine
+    ports:
+      - "8080:80"
+    volumes:
+      - ./nginx.scale.conf:/etc/nginx/conf.d/default.conf:ro
+    depends_on:
+      - api
+
+volumes:
+  pgdata:
+  redisdata:

deploy/nginx.scale.conf

@@ -0,0 +1,29 @@
+# nginx load balancer for the scaled "api" service in docker-compose.scale.yml.
+# Uses Docker's embedded DNS resolver so newly scaled replicas are discovered
+# without editing a static upstream list.
+
+server {
+    listen 80;
+
+    # Docker embedded DNS. valid=10s re-resolves so scaling up/down is picked up.
+    resolver 127.0.0.11 valid=10s;
+
+    location / {
+        # Use a variable so nginx defers resolution to request time (round-robin
+        # across all replicas of the "api" service).
+        set $api_upstream http://api:3001;
+        proxy_pass $api_upstream;
+
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto $scheme;
+
+        # Server-Sent Events: do not buffer the payment status stream.
+        proxy_buffering off;
+        proxy_cache off;
+        proxy_read_timeout 1h;
+        proxy_set_header Connection "";
+        proxy_http_version 1.1;
+    }
+}