docs(env): document GRAPH_BACKEND, Neo4j, and Gemini configuration

Add commented sections for the new pluggable graph backend and Gemini
LLM provider options so users know what variables are available.

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

.env.example

@@ -1,14 +1,31 @@
-# LLM API配置（支持 OpenAI SDK 格式的任意 LLM API）
+# ===== LLM Configuration =====
-# 推荐使用阿里百炼平台qwen-plus模型：https://bailian.console.aliyun.com/
+# Any OpenAI-compatible API is supported
-# 注意消耗较大，可先进行小于40轮的模拟尝试
 LLM_API_KEY=your_api_key_here
 LLM_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
 LLM_MODEL_NAME=qwen-plus
 
-# ===== ZEP记忆图谱配置 =====
+# --- Gemini (Google AI Studio) ---
-# 每月免费额度即可支撑简单使用：https://app.getzep.com/
+# Set LLM_PROVIDER=gemini to auto-configure the Google AI Studio endpoint.
+# Get a free API key at: https://aistudio.google.com/
+# LLM_PROVIDER=gemini
+# LLM_API_KEY=AIza...
+# LLM_MODEL_NAME=gemini-2.0-flash
+
+# ===== Graph Backend =====
+# GRAPH_BACKEND=zep      (default) — Zep Cloud managed memory graph
+# GRAPH_BACKEND=graphiti — Self-hosted Neo4j + Graphiti (requires graphiti extras)
+GRAPH_BACKEND=zep
+
+# --- Zep Cloud (default backend) ---
+# Free tier available: https://app.getzep.com/
 ZEP_API_KEY=your_zep_api_key_here
 
+# --- Graphiti + Neo4j (alternative backend) ---
+# Install extras: pip install "mirofish-backend[graphiti]"
+# NEO4J_URI=bolt://localhost:7687
+# NEO4J_USER=neo4j
+# NEO4J_PASSWORD=your_neo4j_password_here
+
 # ===== 加速 LLM 配置（可选）=====
 # 注意如果不使用加速配置，env文件中就不要出现下面的配置项
 LLM_BOOST_API_KEY=your_api_key_here