后端生成闸+多项修复;前端全站更新;QA 审计与报告

后端:
- 新增 celery_health 生成前置闸——无 worker 在线时图片/视频生成入口
  一律 503,防"提交到 ARK 后无人轮询、结果悬空+额度冻结"的数据丢失
- 拼接导出:帧率跟随源众数、字幕逐句重映射、原声/BGM 混音修复
- 资金核算审计脚本 + genesis 账目回填 + 滞留预留清理命令
- 接入 yunqi provider 与豆包 TTS 模型(catalog/migrations/bootstrap 命令)

前端:全站页面更新(pipeline/library/products/projects/team/account 等),
新增共享 pager 分页组件

QA:刷新 function-audit 全量输出,新增 full-qa 报告
文档:BP 产品介绍资料、design/CLAUDE.md

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

core/backend/apps/ai/providers/yunqi.py

@@ -0,0 +1,43 @@
+from typing import Any
+
+import requests
+from django.conf import settings
+
+from .volcano import VolcanoArkProvider
+
+
+class YunqiProvider(VolcanoArkProvider):
+    """YunQi AI(New API 网关)适配器:OpenAI 兼容 images/generations。
+
+    与火山 ARK 的差异:不接受 watermark/sequential_image_generation/response_format 私有参数;
+    size 用 OpenAI 枚举(1024x1024/1024x1536/1536x1024);gpt-image-2 固定返回 b64_json,
+    下游 extract_first_media_url / media_to_bytes 已兼容 base64,无需改动。
+    """
+
+    def __post_init__(self) -> None:
+        self.api_key = self.api_key or settings.YUNQI.get("api_key")
+        self.base_url = self.base_url or settings.YUNQI.get("base_url")
+
+    def image_generation(
+        self,
+        *,
+        model: str,
+        prompt: str,
+        endpoint: str = "images/generations",
+        image: str | list[str] | None = None,
+        size: str = "1024x1536",
+    ) -> dict[str, Any]:
+        if not self.api_key:
+            raise ValueError("YUNQI_API_KEY is not configured")
+        if image:
+            raise ValueError("gpt-image-2 via YunQi only supports text-to-image; reference image is not supported")
+        body: dict[str, Any] = {"model": model, "prompt": prompt, "size": size, "n": 1}
+        # 实测该网关生图平均延迟 75s+,超时须显著高于火山的 180s
+        response = requests.post(
+            f"{self.base_url.rstrip('/')}/{endpoint.lstrip('/')}",
+            headers={"Authorization": f"Bearer {self.api_key}", "Content-Type": "application/json"},
+            json=body,
+            timeout=300,
+        )
+        response.raise_for_status()
+        return response.json()