feat: 添加MCP调试插件并优化流式超时处理

refactor(relay): 将Provider Key管理路由移至model_config模块
fix(saas): 修复demo_keys与provider_keys的匹配逻辑
perf(runtime): 将流式响应超时从60秒延长至180秒以适配思考型模型
docs: 新增模块化审计和上线前功能测试方案文档
chore: 添加tauri-plugin-mcp依赖及相关配置

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

@@ -5,10 +5,10 @@ pub mod service;
 pub mod handlers;
 pub mod key_pool;
 
-use axum::routing::{delete, get, post, put};
+use axum::routing::{get, post};
 use crate::state::AppState;
 
-/// 中转服务路由 (需要认证)
+/// 中转服务路由 (需要认证 + quota 中间件)
 pub fn routes() -> axum::Router<AppState> {
     axum::Router::new()
         // Relay 核心端点
@@ -17,9 +17,4 @@ pub fn routes() -> axum::Router<AppState> {
         .route("/api/v1/relay/tasks/:id", get(handlers::get_task))
         .route("/api/v1/relay/tasks/:id/retry", post(handlers::retry_task))
         .route("/api/v1/relay/models", get(handlers::list_available_models))
-        // Key Pool 管理 (admin only)
-        .route("/api/v1/providers/:provider_id/keys", get(handlers::list_provider_keys))
-        .route("/api/v1/providers/:provider_id/keys", post(handlers::add_provider_key))
-        .route("/api/v1/providers/:provider_id/keys/:key_id/toggle", put(handlers::toggle_provider_key))
-        .route("/api/v1/providers/:provider_id/keys/:key_id", delete(handlers::delete_provider_key))
 }

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

@@ -15,8 +15,9 @@ use super::types::*;
 /// 上游无数据时，发送 SSE 心跳注释行的间隔
 const STREAMBRIDGE_HEARTBEAT_INTERVAL: Duration = Duration::from_secs(15);
 
-/// 上游无数据时，丢弃连接的超时阈值（90s = 6 个心跳，给 thinking 模型更多时间）
-const STREAMBRIDGE_TIMEOUT: Duration = Duration::from_secs(90);
+/// 上游无数据时，丢弃连接的超时阈值（180s = 12 个心跳）
+/// 实测 Kimi for Coding 的 thinking→content 间隔可达 60s+，需要更宽容的超时。
+const STREAMBRIDGE_TIMEOUT: Duration = Duration::from_secs(180);
 
 /// 流结束后延迟清理的时间窗口
 const STREAMBRIDGE_CLEANUP_DELAY: Duration = Duration::from_secs(60);
@@ -767,9 +768,9 @@ fn build_stream_bridge(
                         idle_heartbeats as u64 * STREAMBRIDGE_HEARTBEAT_INTERVAL.as_secs(),
                     );
 
-                    // After 6 consecutive heartbeats without real data (90s),
+                    // After 12 consecutive heartbeats without real data (180s),
                     // terminate the stream to prevent connection leaks.
-                    if idle_heartbeats >= 6 {
+                    if idle_heartbeats >= 12 {
                         tracing::warn!(
                             "[StreamBridge] Timeout ({:?}) no real data, closing stream for task {}",
                             STREAMBRIDGE_TIMEOUT,