feat: 缓存计费细分、能力匹配优化、用户模型调用计数

1. 缓存创建 tokens 区分 5min/1h TTL，支持按缓存时长差异化计费
   - Usage 表新增 cache_creation_input_tokens_5m/1h 字段
   - Claude handler 解析新格式 (ephemeral_5m/1h, claude_cache_creation_5/1h)
   - 计费规则支持 cache_ttl_pricing 覆盖 cache_creation 价格

2. 能力匹配机制优化
   - COMPATIBLE 能力不再硬过滤，改为排序阶段通过 capability_miss_count 优先级处理
   - cache_1h 改为 COMPATIBLE + REQUEST_PARAM（自动检测请求体中的 ttl=1h）
   - gemini_files 改为 EXCLUSIVE + REQUEST_PARAM（自动检测 fileData.fileUri）
   - 移除前端模型偏好/能力配置 UI（不再需要用户手动配置）

3. 新增用户-模型维度调用次数计数器 (UserModelUsageCount)
   - 原子递增，避免从 Usage 表聚合查询
   - 前端模型目录和用户可用模型列表展示调用次数

4. 其他改进
   - global_model_id 改为必填（NOT NULL），清理孤立模型
   - 模型映射对话框支持从上游获取模型列表并分组折叠
   - 端点测试不再依赖端点启用状态
   - 异步任务页面对普通用户隐藏用户信息列
   - Dashboard 响应式布局断点调整 (sm -> lg)
   - 号池管理仅展示已启用号池的提供商

src/api/admin/provider_query.py

@@ -30,6 +30,7 @@ from src.services.model.fetch_scheduler import (
 from src.services.model.upstream_fetcher import (
     EndpointFetchConfig,
     UpstreamModelsFetchContext,
+    UpstreamModelsFetcherRegistry,
     build_format_to_config,
     fetch_models_for_key,
     get_adapter_for_format,
@@ -235,7 +236,15 @@ async def query_available_models(
     # 构建 api_format -> EndpointFetchConfig 映射（纯数据，不依赖 ORM session）
     format_to_endpoint = build_format_to_config(provider.endpoints)
 
-    if not format_to_endpoint:
+    # 检查是否有注册自定义 fetcher（如预设模型），有则不依赖活跃 endpoint
+    provider_type = str(getattr(provider, "provider_type", "") or "").lower()
+    # 延迟导入避免循环依赖（与 upstream_fetcher.fetch_models_for_key 保持一致）
+    from src.services.provider.envelope import ensure_providers_bootstrapped
+
+    ensure_providers_bootstrapped()
+    has_custom_fetcher = UpstreamModelsFetcherRegistry.get(provider_type) is not None
+
+    if not format_to_endpoint and not has_custom_fetcher:
         raise HTTPException(status_code=400, detail="No active endpoints found for this provider")
 
     # 如果指定了 api_key_id，只获取该 Key 的模型
@@ -253,7 +262,6 @@ async def query_available_models(
         raise HTTPException(status_code=400, detail="No active API Key found for this provider")
 
     # Antigravity: 按 tier/可用性排序后逐个尝试，成功即停止
-    provider_type = str(getattr(provider, "provider_type", "") or "").lower()
     if provider_type == ProviderType.ANTIGRAVITY:
         return await _fetch_models_antigravity_ordered(
             provider=provider,
@@ -610,12 +618,12 @@ async def test_model(
         raise HTTPException(status_code=404, detail="Provider not found")
 
     # 构建 api_format -> endpoint 映射 和 id -> endpoint 映射
+    # 测试不依赖端点启用状态，禁用的端点也可以用于测试连通性
     format_to_endpoint: dict[str, ProviderEndpoint] = {}
     id_to_endpoint: dict[str, ProviderEndpoint] = {}
     for ep in provider.endpoints:
-        if ep.is_active:
-            format_to_endpoint[ep.api_format] = ep
-            id_to_endpoint[ep.id] = ep
+        format_to_endpoint[ep.api_format] = ep
+        id_to_endpoint[ep.id] = ep
 
     # 找到合适的端点和 API Key
     endpoint = None
@@ -628,7 +636,7 @@ async def test_model(
         if not endpoint:
             raise HTTPException(
                 status_code=404,
-                detail=f"No active endpoint found for API format: {request.api_format}",
+                detail=f"No endpoint found for API format: {request.api_format}",
             )
 
         if request.api_key_id:
@@ -657,7 +665,7 @@ async def test_model(
         # 使用指定的端点
         endpoint = id_to_endpoint.get(request.endpoint_id)
         if not endpoint:
-            raise HTTPException(status_code=404, detail="Endpoint not found or not active")
+            raise HTTPException(status_code=404, detail="Endpoint not found")
 
         if request.api_key_id:
             # 同时指定了 Key，需要校验是否支持该端点格式

src/api/admin/providers/models.py

@@ -715,6 +715,7 @@ class AdminImportFromUpstreamAdapter(AdminApiAdapter):
                     # 1. 检查是否已存在同名的 ProviderModel
                     existing = (
                         db.query(Model)
+                        .options(joinedload(Model.global_model))
                         .filter(
                             Model.provider_id == self.provider_id,
                             Model.provider_model_name == model_id,
@@ -727,10 +728,8 @@ class AdminImportFromUpstreamAdapter(AdminApiAdapter):
                         success.append(
                             ImportFromUpstreamSuccessItem(
                                 model_id=model_id,
-                                global_model_id=existing.global_model_id or "",
-                                global_model_name=(
-                                    existing.global_model.name if existing.global_model else ""
-                                ),
+                                global_model_id=existing.global_model_id,
+                                global_model_name=existing.global_model.name,
                                 provider_model_id=existing.id,
                                 created_global_model=False,
                             )

src/api/admin/providers/summary.py

@@ -319,7 +319,6 @@ def _build_provider_summary(db: Session, provider: Provider) -> ProviderWithEndp
         .filter(
             Model.provider_id == provider.id,
             Model.is_active == True,
-            Model.global_model_id.isnot(None),
         )
         .distinct()
         .all()

src/api/admin/system.py

@@ -2556,16 +2556,18 @@ def _purge_stats_and_reset_counters(db: Session) -> None:
 class AdminPurgeUsageAdapter(AdminApiAdapter):
     async def handle(self, context: ApiRequestContext) -> Any:  # type: ignore[override]
         """清空全部使用记录及相关统计数据"""
-        from src.models.database import RequestCandidate
+        from src.models.database import RequestCandidate, UserModelUsageCount
 
         db = context.db
 
         usage_count = db.query(Usage).count()
         candidates_count = db.query(RequestCandidate).count()
+        usage_counts_count = db.query(UserModelUsageCount).count()
 
         # 清空使用记录
         db.query(RequestCandidate).delete()
         db.query(Usage).delete()
+        db.query(UserModelUsageCount).delete()
 
         _purge_stats_and_reset_counters(db)
         db.commit()
@@ -2575,6 +2577,7 @@ class AdminPurgeUsageAdapter(AdminApiAdapter):
             "deleted": {
                 "usage_records": usage_count,
                 "request_candidates": candidates_count,
+                "user_model_usage_counts": usage_counts_count,
             },
         }
 

src/api/admin/usage/routes.py

@@ -1259,6 +1259,8 @@ class AdminUsageDetailAdapter(AdminApiAdapter):
             },
             "cache_creation_input_tokens": usage_record.cache_creation_input_tokens,
             "cache_read_input_tokens": usage_record.cache_read_input_tokens,
+            "cache_creation_input_tokens_5m": usage_record.cache_creation_input_tokens_5m or 0,
+            "cache_creation_input_tokens_1h": usage_record.cache_creation_input_tokens_1h or 0,
             "cache_creation_cost": getattr(usage_record, "cache_creation_cost_usd", 0.0),
             "cache_read_cost": getattr(usage_record, "cache_read_cost_usd", 0.0),
             "request_cost": getattr(usage_record, "request_cost_usd", 0.0),