refactor(crates): kernel/generation module split + DeerFlow optimizations + middleware + dead code cleanup

- Split zclaw-kernel/kernel.rs (1486 lines) into 9 domain modules
- Split zclaw-kernel/generation.rs (1080 lines) into 3 modules
- Add DeerFlow-inspired middleware: DanglingTool, SubagentLimit, ToolError, ToolOutputGuard
- Add PromptBuilder for structured system prompt assembly
- Add FactStore (zclaw-memory) for persistent fact extraction
- Add task builtin tool for agent task management
- Driver improvements: Anthropic/OpenAI extended thinking, Gemini safety settings
- Replace let _ = with proper log::warn! across SaaS handlers
- Remove unused dependency (url) from zclaw-hands

crates/zclaw-saas/src/relay/service.rs

@@ -259,7 +259,9 @@ pub async fn execute_relay(
             }
         }
 
-        let key_id = current_key_id.as_ref().unwrap().clone();
+        let key_id = current_key_id.as_ref()
+            .ok_or_else(|| SaasError::Internal("Key pool selection failed: no key_id".into()))?
+            .clone();
         let api_key = current_api_key.clone();
 
         let mut req_builder = client.post(&url)
@@ -309,7 +311,10 @@ pub async fn execute_relay(
                                     }
                                 }
                                 Err(e) => {
-                                    let _ = tx.send(Err(std::io::Error::other(e))).await;
+                                    let err_msg = e.to_string();
+                                    if tx.send(Err(std::io::Error::other(e))).await.is_err() {
+                                        tracing::debug!("SSE relay: client disconnected before error sent: {}", err_msg);
+                                    }
                                     break;
                                 }
                             }
@@ -372,12 +377,12 @@ pub async fn execute_relay(
                     let (input_tokens, output_tokens) = extract_token_usage(&body);
                     update_task_status(db, task_id, "completed",
                         Some(input_tokens), Some(output_tokens), None).await?;
-                    // 记录 Key 使用量
-                    let _ = super::key_pool::record_key_usage(
+                    // 记录 Key 使用量（失败仅记录，不阻塞响应）
+                    if let Err(e) = super::key_pool::record_key_usage(
                         db, &key_id, Some(input_tokens + output_tokens),
-                    ).await.map_err(|e| {
+                    ).await {
                         tracing::warn!("[Relay] Failed to record key usage for billing: {}", e);
-                    });
+                    }
                     return Ok(RelayResponse::Json(body));
                 }
             }
@@ -557,7 +562,10 @@ fn hash_request(body: &str) -> String {
 fn extract_token_usage(body: &str) -> (i64, i64) {
     let parsed: serde_json::Value = match serde_json::from_str(body) {
         Ok(v) => v,
-        Err(_) => return (0, 0),
+        Err(e) => {
+            tracing::debug!("extract_token_usage: JSON parse failed (body len={}): {}", body.len(), e);
+            return (0, 0);
+        }
     };
 
     let usage = parsed.get("usage");