feat: add Suyao Feishu bot and per-agent memory config support

- Create suyao_app_service.py and suyao_ws_handler.py for 苏瑶 Feishu bot
- Add SUYAO_APP_ID/SUYAO_APP_SECRET/SUYAO_AGENT_ID config fields
- Fix node config extraction bug (n.get("config") → n.get("data")) in all WS handlers
- Add _build_memory_config_from_node() to support per-agent memory settings
  (max_history_messages, vector_memory_top_k, persist_to_db, etc.)
- Create 苏瑶1号 (Plan A: long context), 苏瑶2号 (Plan B: emotion tracking),
  苏瑶3号 (Plan C: knowledge graph + RAG) agents with different memory strategies

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

backend/app/services/feishu_ws_handler.py

@@ -153,7 +153,7 @@ async def _handle_message_async(data):
 
         _reply_to_feishu(open_id, f"🤔 正在思考，请稍候...")
 
-        from app.agent_runtime import AgentRuntime, AgentConfig, AgentLLMConfig, AgentToolConfig
+        from app.agent_runtime import AgentRuntime, AgentConfig, AgentLLMConfig, AgentToolConfig, AgentMemoryConfig
 
         wc = agent.workflow_config or {}
         nodes = wc.get("nodes", [])
@@ -164,14 +164,15 @@ async def _handle_message_async(data):
         max_iterations = 10
 
         for n in nodes:
-            cfg = n.get("config", {}) if isinstance(n, dict) else getattr(n, "config", {})
-            if cfg.get("type") in ("agent", "llm"):
-                system_prompt = cfg.get("system_prompt", "") or system_prompt
-                model = cfg.get("model", model)
-                provider = cfg.get("provider", provider)
-                temperature = float(cfg.get("temperature", temperature))
-                max_iterations = int(cfg.get("max_iterations", max_iterations))
-                break
+            if n.get("type") not in ("agent", "llm", "template"):
+                continue
+            cfg = n.get("data", {}) if isinstance(n, dict) else getattr(n, "data", {})
+            system_prompt = cfg.get("system_prompt", "") or system_prompt
+            model = cfg.get("model", model)
+            provider = cfg.get("provider", provider)
+            temperature = float(cfg.get("temperature", temperature))
+            max_iterations = int(cfg.get("max_iterations", max_iterations))
+            break
 
         config = AgentConfig(
             name=agent.name or "agent",
@@ -183,6 +184,13 @@ async def _handle_message_async(data):
                 max_iterations=max_iterations,
             ),
             tools=AgentToolConfig(),
+            memory=AgentMemoryConfig(
+                max_history_messages=int(cfg.get("memory_max_history", 20)),
+                vector_memory_top_k=int(cfg.get("memory_vector_top_k", 5)),
+                persist_to_db=bool(cfg.get("memory_persist", True)),
+                vector_memory_enabled=bool(cfg.get("memory_vector_enabled", True)),
+                learning_enabled=bool(cfg.get("memory_learning", True)),
+            ),
             user_id=user.id,
             memory_scope_id=str(agent.id),
         )