fix: 性能优化、前端转写器配置、任务进度丢失及 MLX Whisper 回退问题修复

### 性能优化
- 后端任务执行从串行锁改为 ThreadPoolExecutor 并发执行（默认3线程）
- 添加 GZipMiddleware 响应压缩 + Nginx gzip 配置
- 数据库连接池参数优化（pool_size=10, max_overflow=20）
- 视频帧提取并行化（ThreadPoolExecutor）
- LLM 重试配置缓存到实例，避免每次请求读 env var
- 前端路由级代码拆分（React.lazy + Suspense）
- Vite manualChunks 拆分 markdown/markmap/vendor
- MarkdownViewer 用 React.memo + useMemo 减少不必要渲染
- NoteHistory Fuse.js 实例 useMemo 缓存
- useTaskPolling 无待处理任务时跳过轮询
- 移除 antd 依赖（NoteForm Alert、modelForm Tag），改用 shadcn/ui

### 前端转写器配置（新功能）
- 新增 TranscriberConfigManager（JSON 文件存储，替代环境变量）
- 新增 GET/POST /transcriber_config API 端点
- 新增 GET /transcriber_models_status 模型下载状态查询
- 新增 POST /transcriber_download 后台模型下载触发
- 前端转写器设置页面：引擎选择、模型大小选择、模型下载管理
- deploy_status 端点同步从配置文件读取

### Bug 修复
- 修复任务进行中切换页面后进度丢失：Home.tsx status 派生逻辑补全中间状态
- 修复 MLX Whisper 静默回退 fast-whisper：移除环境变量门控，macOS 下自动尝试导入
- MLX Whisper 不可用时抛出 RuntimeError 而非静默回退
- 前端展示 MLX Whisper 可用性状态，不可用时禁用保存

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

backend/app/routers/config.py

@@ -1,13 +1,23 @@
-from fastapi import APIRouter, HTTPException
+import os
+import platform
+from pathlib import Path
+
+from fastapi import APIRouter, HTTPException, BackgroundTasks
 from pydantic import BaseModel
 from typing import Optional
 from app.utils.response import ResponseWrapper as R
+from app.utils.logger import get_logger
+from app.utils.path_helper import get_model_dir
 
 from app.services.cookie_manager import CookieConfigManager
+from app.services.transcriber_config_manager import TranscriberConfigManager
 from ffmpeg_helper import ensure_ffmpeg_or_raise
 
+logger = get_logger(__name__)
+
 router = APIRouter()
 cookie_manager = CookieConfigManager()
+transcriber_config_manager = TranscriberConfigManager()
 
 
 class CookieUpdateRequest(BaseModel):
@@ -32,6 +42,165 @@ def update_cookie(data: CookieUpdateRequest):
 
     )
 
+class TranscriberConfigRequest(BaseModel):
+    transcriber_type: str
+    whisper_model_size: Optional[str] = None
+
+
+AVAILABLE_TRANSCRIBER_TYPES = [
+    {"value": "fast-whisper", "label": "Faster Whisper（本地）"},
+    {"value": "bcut", "label": "必剪（在线）"},
+    {"value": "kuaishou", "label": "快手（在线）"},
+    {"value": "groq", "label": "Groq（在线）"},
+    {"value": "mlx-whisper", "label": "MLX Whisper（仅macOS）"},
+]
+
+WHISPER_MODEL_SIZES = ["tiny", "base", "small", "medium", "large-v3", "large-v3-turbo"]
+
+
+@router.get("/transcriber_config")
+def get_transcriber_config():
+    from app.transcriber.transcriber_provider import MLX_WHISPER_AVAILABLE
+
+    config = transcriber_config_manager.get_config()
+    return R.success(data={
+        **config,
+        "available_types": AVAILABLE_TRANSCRIBER_TYPES,
+        "whisper_model_sizes": WHISPER_MODEL_SIZES,
+        "mlx_whisper_available": MLX_WHISPER_AVAILABLE,
+    })
+
+
+@router.post("/transcriber_config")
+def update_transcriber_config(data: TranscriberConfigRequest):
+    config = transcriber_config_manager.update_config(
+        transcriber_type=data.transcriber_type,
+        whisper_model_size=data.whisper_model_size,
+    )
+    return R.success(data=config)
+
+
+# ---- Whisper 模型下载状态 & 下载触发 ----
+
+# 用于跟踪正在进行的下载任务
+_downloading: dict[str, str] = {}  # model_size -> status ("downloading" | "done" | "failed")
+
+
+def _check_whisper_model_exists(model_size: str, subdir: str = "whisper") -> bool:
+    """检查指定 whisper 模型是否已下载到本地。"""
+    model_dir = get_model_dir(subdir)
+    model_path = os.path.join(model_dir, f"whisper-{model_size}")
+    return Path(model_path).exists()
+
+
+@router.get("/transcriber_models_status")
+def get_transcriber_models_status():
+    """返回所有 whisper 模型的下载状态。"""
+    statuses = []
+    for size in WHISPER_MODEL_SIZES:
+        downloaded = _check_whisper_model_exists(size, "whisper")
+        download_status = _downloading.get(size)
+        statuses.append({
+            "model_size": size,
+            "downloaded": downloaded,
+            "downloading": download_status == "downloading",
+        })
+
+    # 也检查 mlx-whisper（仅 macOS）
+    mlx_available = platform.system() == "Darwin"
+    mlx_statuses = []
+    if mlx_available:
+        for size in WHISPER_MODEL_SIZES:
+            mlx_key = f"mlx-{size}"
+            model_dir = get_model_dir("mlx-whisper")
+            model_path = os.path.join(model_dir, f"mlx-community/whisper-{size}")
+            downloaded = Path(model_path).exists()
+            mlx_statuses.append({
+                "model_size": size,
+                "downloaded": downloaded,
+                "downloading": _downloading.get(mlx_key) == "downloading",
+            })
+
+    return R.success(data={
+        "whisper": statuses,
+        "mlx_whisper": mlx_statuses,
+        "mlx_available": mlx_available,
+    })
+
+
+class ModelDownloadRequest(BaseModel):
+    model_size: str
+    transcriber_type: str = "fast-whisper"  # "fast-whisper" 或 "mlx-whisper"
+
+
+def _do_download_whisper(model_size: str):
+    """后台下载 faster-whisper 模型。"""
+    from app.transcriber.whisper import MODEL_MAP
+    from modelscope import snapshot_download
+
+    try:
+        _downloading[model_size] = "downloading"
+        model_dir = get_model_dir("whisper")
+        model_path = os.path.join(model_dir, f"whisper-{model_size}")
+        if Path(model_path).exists():
+            _downloading[model_size] = "done"
+            return
+        repo_id = MODEL_MAP.get(model_size)
+        if not repo_id:
+            _downloading[model_size] = "failed"
+            return
+        logger.info(f"开始下载 whisper 模型: {model_size}")
+        snapshot_download(repo_id, local_dir=model_path)
+        logger.info(f"whisper 模型下载完成: {model_size}")
+        _downloading[model_size] = "done"
+    except Exception as e:
+        logger.error(f"whisper 模型下载失败: {model_size}, {e}")
+        _downloading[model_size] = "failed"
+
+
+def _do_download_mlx_whisper(model_size: str):
+    """后台下载 mlx-whisper 模型。"""
+    key = f"mlx-{model_size}"
+    try:
+        _downloading[key] = "downloading"
+        from huggingface_hub import snapshot_download as hf_download
+
+        model_dir = get_model_dir("mlx-whisper")
+        model_name = f"mlx-community/whisper-{model_size}"
+        model_path = os.path.join(model_dir, model_name)
+        if Path(model_path).exists():
+            _downloading[key] = "done"
+            return
+        logger.info(f"开始下载 mlx-whisper 模型: {model_size}")
+        hf_download(model_name, local_dir=model_path, local_dir_use_symlinks=False)
+        logger.info(f"mlx-whisper 模型下载完成: {model_size}")
+        _downloading[key] = "done"
+    except Exception as e:
+        logger.error(f"mlx-whisper 模型下载失败: {model_size}, {e}")
+        _downloading[key] = "failed"
+
+
+@router.post("/transcriber_download")
+def download_transcriber_model(data: ModelDownloadRequest, background_tasks: BackgroundTasks):
+    """触发后台下载指定的 whisper 模型。"""
+    if data.model_size not in WHISPER_MODEL_SIZES:
+        return R.error(msg=f"不支持的模型大小: {data.model_size}")
+
+    if data.transcriber_type == "mlx-whisper":
+        if platform.system() != "Darwin":
+            return R.error(msg="MLX Whisper 仅支持 macOS")
+        key = f"mlx-{data.model_size}"
+        if _downloading.get(key) == "downloading":
+            return R.success(msg="模型正在下载中")
+        background_tasks.add_task(_do_download_mlx_whisper, data.model_size)
+    else:
+        if _downloading.get(data.model_size) == "downloading":
+            return R.success(msg="模型正在下载中")
+        background_tasks.add_task(_do_download_whisper, data.model_size)
+
+    return R.success(msg="模型下载已开始")
+
+
 @router.get("/sys_health")
 async def sys_health():
     try:
@@ -59,9 +228,10 @@ async def deploy_status():
         "gpu_name": torch.cuda.get_device_name(0) if cuda_available else None,
     }
     
-    # Whisper 模型状态
-    model_size = os.getenv("WHISPER_MODEL_SIZE", "base")
-    transcriber_type = os.getenv("TRANSCRIBER_TYPE", "fast-whisper")
+    # Whisper 模型状态（从配置文件读取，与前端设置同步）
+    transcriber_cfg = transcriber_config_manager.get_config()
+    model_size = transcriber_cfg["whisper_model_size"]
+    transcriber_type = transcriber_cfg["transcriber_type"]
     
     # FFmpeg 状态
     try:

backend/app/routers/note.py

@@ -101,7 +101,7 @@ def run_note_task(task_id: str, video_url: str, platform: str, quality: Download
             grid_size=grid_size,
         )
 
-    logger.info(f"任务进入串行队列，等待执行 (task_id={task_id})")
+    logger.info(f"任务进入执行队列 (task_id={task_id})")
     note = task_serial_executor.run(_execute_note_task)
     logger.info(f"Note generated: {task_id}")
     if not note or not note.markdown: