chore: 干净 ERP 基座 — 删除 health/ai/wechat 业务代码

删除内容:
- 前端: health/(67文件), ai/(2文件), Copilot, MediaPicker, 相关API/Store/Hook
- 后端: wechat_handler, wechat_service, wechat_user entity, analytics handler, ai_workflow_seed
- 配置: WechatConfig, AppConfig.wechat, AuthState wechat 字段
- 启动: 微信凭据检查块, ensure_ai_workflows() 调用
- 迁移: 新增 m20260613_000170_drop_wechat_users.rs
- 脚本: api_test_health_alert.py, api_test_mp.py, mpsync.sh/ps1
- E2E: health-data page, flows/ 目录

保留: erp-core/auth/workflow/message/config/plugin + 基座前端 + 通用组件

docker/docker-compose.cloud.yml

@@ -0,0 +1,40 @@
+# 云端部署配置 — 仅启动应用容器，PG/Redis 使用宿主机已安装的服务
+# 使用方式: docker compose -f docker/docker-compose.cloud.yml up -d
+#
+# 前置条件:
+#   1. 宿主机已安装 PostgreSQL 16 + Redis 7
+#   2. PostgreSQL 已创建数据库和用户
+#   3. 复制 .env.production.example 为 .env.production 并填写实际值
+#   4. OpenResty 反代配置:
+#      - /api/* → http://localhost:3000
+#      - /uploads/* → http://localhost:3000
+#      - / → 前端静态文件 (挂载 /opt/hms/static/)
+
+services:
+  app:
+    build:
+      context: ..
+      dockerfile: Dockerfile
+    container_name: hms-server
+    restart: unless-stopped
+    network_mode: host
+    env_file:
+      - .env.production
+    volumes:
+      - ../uploads:/app/uploads
+      - ../config:/app/config:ro
+      - ../static:/app/static
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:3000/api/v1/health"]
+      interval: 30s
+      timeout: 5s
+      start_period: 60s
+      retries: 3
+    deploy:
+      resources:
+        limits:
+          cpus: "2"
+          memory: 1024M
+        reservations:
+          cpus: "0.5"
+          memory: 256M