Implement three divination modes, learn pages, and PM2 deploy on port 3130.

Add liuyao/bazi/combined flows with shared calc and AI infrastructure, 64-gua learn routes, and update Ubuntu PM2 deployment docs for port 3130.

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

.env.example

@@ -13,8 +13,8 @@ OPENAI_BASE_URL=https://op.bz121.com/v1
 # AI 模型
 OPENAI_MODEL=huihui_ai/gemma-4-abliterated:e4b
 
-# 服务端口
-PORT=3000
+# 服务端口（生产 PM2 默认 3130；本地 dev 仍用 3000）
+PORT=3130
 
 # 运行环境
 NODE_ENV=production