feat(proxy): 架构与算法优化二轮——sense 运行时单例激活 T-X3、打码密钥泄漏修复、账本热路径 2.4x

- gateway/proxy/routes.py：sense 运行时（SenseStore+Grader+晋升表）按配置单例化，
  原每请求重建导致：每请求跑全量 DDL、重读工件文件，且 T-X3 60s 决策缓存随
  Grader 丢弃、命中率恒为 0（本轮最大收益，激活既有已测组件）；
  T1 审计抽样修复原 NameError×3 被 except 吞掉（load_config/get_review/request_id
  均未定义、§9.4 从未入队）——request_id 上提、review 队列经 review_getter 注入
  （api.py 反向依赖解除）、sample_rate 走 load_config
- gateway/model_pool.py：新增 usable_entries() 未打码内通道；routes 的降级链/
  预算降档/能力位重定向/档位映射四处改走该通道——修复打码 api_key 流入上游
  Bearer 头导致带密钥条目重定向必 401 的隐性缺陷（HTTP 管理面仍用打码 list()）
- gateway/proxy/ledger.py + billing.py：共享持久连接（原每操作新建，全链路
  每请求 3-5 次 connect）+ 日重置 UPDATE 每实例每日一次短路（原每请求全表扫描
  抢写锁）；check_and_count A/B 0.75→0.31 ms/op（2.4x）；新增 usage_stats()
  SQL 聚合，/admin/stats 从 list_usage(limit=50 万) Python 四遍扫描改为下推聚合
  （微基准 ~28x，随流水线性扩大）
- gateway/sense/grader.py：决策留痕 insert_decision 移入 asyncio.to_thread
  （原同步 sqlite 写直接跑在事件循环线程，高并发阻塞网关；与观察写批量队列同等保护）
- 全量 474 项两轮复核全绿（基线 474）

gateway/proxy/billing.py

@@ -179,6 +179,39 @@ class BillingMixin:
                 (request_id,)).fetchone()
         return dict(row) if row else None
 
+    def usage_stats(self) -> Dict[str, Any]:
+        """全局流水聚合（§5.2 看板口径，2026-09 优化：SQL 下推替代全量拉取）。
+
+        返回 {requests, gateway_cached, in_hit_tok, in_miss_tok, revenue_milli,
+        cost_milli, by_bucket}；by_bucket 键为 bucket 名。
+        """
+        with self._lock, self._connect() as conn:
+            row = conn.execute(
+                "SELECT COUNT(*) AS requests,"
+                " COALESCE(SUM(gateway_cached), 0) AS gateway_cached,"
+                " COALESCE(SUM(in_hit_tok), 0) AS in_hit_tok,"
+                " COALESCE(SUM(in_miss_tok), 0) AS in_miss_tok,"
+                " COALESCE(SUM(charged_milli), 0) AS revenue_milli,"
+                " COALESCE(SUM(upstream_cost_milli), 0) AS cost_milli"
+                " FROM usage_ledger").fetchone()
+            buckets = conn.execute(
+                "SELECT bucket, COUNT(*) AS requests,"
+                " COALESCE(SUM(charged_milli), 0) AS revenue_milli,"
+                " COALESCE(SUM(upstream_cost_milli), 0) AS cost_milli"
+                " FROM usage_ledger GROUP BY bucket").fetchall()
+        return {
+            "requests": row["requests"],
+            "gateway_cached": row["gateway_cached"],
+            "in_hit_tok": row["in_hit_tok"],
+            "in_miss_tok": row["in_miss_tok"],
+            "revenue_milli": row["revenue_milli"],
+            "cost_milli": row["cost_milli"],
+            "by_bucket": {r["bucket"]: {"requests": r["requests"],
+                                        "revenue_milli": r["revenue_milli"],
+                                        "cost_milli": r["cost_milli"]}
+                          for r in buckets},
+        }
+
     def list_usage(self, student_id: Optional[int] = None,
                    limit: int = 50, offset: int = 0) -> List[Dict[str, Any]]:
         """流水分页（可按学生过滤，经其名下 key）；按学生过滤走联表常量语句。"""