fix(optimize): 参数优化端到端修复(空statistics/500/不跳页)

根因: vnpy run_optimization 孙进程 spawn re-import setting.py 重置 DB → load 0根 → 空 statistics;连带 task_id 不一致 + statistics 含 date/numpy 序列化失败 + 前端单页轮询 180s 超时。
修复(不动 vnpy 源码):
- cta_optimizer: vt_setting.json 适配(vnpy 原生机制 setting.py:43 load_json,孙进程拿到正确 DB) + task_id 贯通 + 主聚合持久化(combos) + 过滤空 statistics 记录
- result_store: _json_default(date→isoformat + numpy→原生),save_result 两处 json.dumps 加 default
- routes: optimization-results 改读 DB 主聚合, fallback 内存(重启不丢)
- runner: _opt_worker 加 task_id 参数 + submit_optimize 传递(对齐 _cta_worker)
- 前端: Optimize 提交后跳 progress;Progress 用 opt_ 前缀判断跳 optimize-result;新建 OptimizeResult 参数表格页;router 加路由
验证: 端到端 optimization-results http=200 + 4组合非空 statistics + 浏览器提交→跳 optimize-result 页表格渲染。

sanguo_orchestrator/runner.py

@@ -82,7 +82,7 @@ class Orchestrator:
         spec = self._pending[task_id]
         fut: Future = self.pool.submit_work(
             task_id, _opt_worker, spec["strategy_class"], spec["symbol"],
-            spec["grid"], spec["start"], spec["end"], spec["cfg"], self.db_path
+            spec["grid"], spec["start"], spec["end"], spec["cfg"], self.db_path, task_id
         )
 
         task = self.pool.get_task(task_id)
@@ -178,10 +178,10 @@ def _cta_worker(strategy_class, symbol: str, params: dict, start: str, end: str,
     return run_cta_backtest(strategy_class, symbol, params, start, end, cfg, db_path, benchmark=benchmark, task_id=task_id, capital=capital, position_pct=position_pct)
 
 
-def _opt_worker(strategy_class, symbol: str, grid: dict, start: str, end: str, cfg, db_path: str) -> any:
+def _opt_worker(strategy_class, symbol: str, grid: dict, start: str, end: str, cfg, db_path: str, task_id: str) -> any:
     """Worker for CTA optimization (lazy import, spawn-friendly)"""
     from sanguo_backtest.cta_optimizer import run_cta_optimization
-    return run_cta_optimization(strategy_class, symbol, grid, start, end, cfg, db_path)
+    return run_cta_optimization(strategy_class, symbol, grid, start, end, cfg, db_path, task_id=task_id)
 
 
 def _factor_worker(symbols: list, factor_names: list, start: str, end: str, cfg, output_dir: str) -> any: