[架构重构] 移除本地Ollama融合，云端直出JSON直存DB，Q&A三模型降级

1. 视频摘要链路：云端VLM直出结构化JSON → Edge format_cloud_result格式化校验 → 直存NAS DB（移除run_text_fusion本地融合）
2. 智能问答链路：Gemini→NVIDIA→Ollama降级，新增chat()纯文本问答方法
3. 适配器重构：base/gemini/nvidia/ollama adapter新增chat()；gemini多图单请求结构化JSON；nvidia逐帧调用聚合
4. 端点变更：/api/edge/chat → /api/edge/chat/ask，调orchestrator.run_qa()
5. chat_handler改经Edge Q&A编排，不再直连Ollama
6. 配置更新：ollama_url → qa_url，Ollama role注释改为Q&A兜底
7. README同步更新架构描述、拓扑图、时序图、模块表

fam-core/src/fam_core/chat_handler/chat_handler.py

@@ -71,27 +71,25 @@ def _aggregate_by_hour(details):
     return '\n'.join(lines)
 
 
-def _call_ollama(prompt: str) -> str:
-    """调用 Oracle Ollama 纯文本模式"""
+def _call_edge_qa(prompt: str) -> str:
+    """调用 FAM-Edge 问答编排端点（Gemini → NVIDIA → 本地 Ollama 兜底）"""
     cfg = load_config()
-    ollama_url = cfg.get('chat_handler', {}).get(
-        'ollama_url', 'http://localhost:11434/api/generate'
+    qa_url = cfg.get('chat_handler', {}).get(
+        'qa_url', 'http://129.146.203.203:5000/api/edge/chat/ask'
     )
-    model_name = cfg.get('chat_handler', {}).get('model_name', 'llava-phi3')
     timeout = cfg.get('chat_handler', {}).get('timeout', 120)
 
-    resp = requests.post(ollama_url, json={
-        "model": model_name,
-        "prompt": prompt,
-        "stream": False,
-        "options": {"temperature": 0.3}
-    }, timeout=timeout)
+    resp = requests.post(qa_url, json={"prompt": prompt}, timeout=timeout)
 
     if resp.status_code == 200:
-        return resp.json().get('response', '')
+        data = resp.json()
+        answer = data.get('answer', '')
+        provider = data.get('provider', 'unknown')
+        logger.info(f"问答由 {provider} 提供回答（长度={len(answer)}）")
+        return answer
     else:
-        logger.error(f"Ollama 调用失败: {resp.status_code} {resp.text}")
-        raise Exception(f"Ollama error: {resp.status_code}")
+        logger.error(f"Edge 问答调用失败: {resp.status_code} {resp.text[:200]}")
+        raise Exception(f"Edge QA error: {resp.status_code}")
 
 
 @chat_bp.route('/api/chat/ask', methods=['POST'])
@@ -136,9 +134,9 @@ def chat_ask():
         )
 
         try:
-            answer = _call_ollama(prompt)
+            answer = _call_edge_qa(prompt)
         except Exception as e:
-            logger.error(f"Ollama 调用失败: {e}")
+            logger.error(f"问答编排调用失败: {e}")
             return jsonify({"error": f"AI 调用失败: {e}"}), 503
 
     # 3. 写入 chat_history