feat(proxy,billing,pool): 增强隧道/流中断诊断日志、修复缓存 TTL 差异化计价

- stream_processor: 上游流中断时记录完整异常链与已传输 token 统计
- hub_transport: 断连影响 in-flight 流、STREAM_ERROR、超时场景补充 warning 日志；
  未启用时跳过重连循环，重连失败日志降频避免刷屏
- tunnel_manager: 流超时/错误/全部取消/STREAM_ERROR 增加诊断日志与字节统计
- billing_integration: 计费时自动补全 cache_ttl_minutes（从 provider key 查询
  或从 5m/1h 细分 token 回推），修复缓存 TTL 差异化计价缺失
- PoolManagement.vue: 移除重复的账号告警 Badge
- 新增 billing integration 单元测试

src/api/handlers/base/stream_processor.py

@@ -1057,6 +1057,31 @@ class StreamProcessor:
         except GeneratorExit:
             raise
         except (httpx.StreamClosed, httpx.HTTPError) as exc:
+            # 记录上游流中断的详细诊断信息
+            elapsed = time.monotonic() - start_time if start_time else 0
+            exc_chain = []
+            seen: set[int] = set()
+            cause: BaseException | None = exc
+            while cause is not None and id(cause) not in seen:
+                seen.add(id(cause))
+                exc_chain.append(f"{type(cause).__name__}: {cause}")
+                cause = cause.__cause__ or cause.__context__
+            logger.warning(
+                "[{}] upstream stream error: provider={}, model={}, "
+                "yielded_any={}, has_completion={}, elapsed={:.1f}s, "
+                "input_tokens={}, output_tokens={}, "
+                "exception_chain=[{}]",
+                self.request_id,
+                ctx.provider_name,
+                ctx.model,
+                yielded_any,
+                ctx.has_completion,
+                elapsed,
+                ctx.input_tokens,
+                ctx.output_tokens,
+                " -> ".join(exc_chain),
+            )
+
             # 连接关闭/协议错误：best-effort flush 残留 SSE，避免丢失尾部 usage。
             try:
                 if buffer: