feat: 视频调速引擎（快进/慢放）

- SpeedEngine 抽象层：SpeedConfig + SpeedEngine，支持 0.25x~4x 变速
- 视频调速：基于 FFmpeg setpts，在 clip 预处理环节插入
- 音频调速：基于 atempo，超范围自动多级串联（如 4x=2.0*2.0）
- 音画同步：视频音频同时调速，音调自动修正
- 分段调速：每个 clip 独立 playback_speed 字段
- 整段调速：所有 clip 设相同 speed 即可实现
- 降级策略：速度超出范围自动钳制，不阻断渲染
- 领域模型：EditPlanClip 新增 playback_speed 字段
- 数据库：EditPlanClipModel 新增 playback_speed 列
- API：create_clip / update_clip 支持 playback_speed 参数
- 45 个新增单测全绿 + 100 个现有测试全绿，无回归

apps/worker/video_processing/render_audio.py

@@ -16,6 +16,8 @@ import subprocess
 from dataclasses import dataclass, field
 from pathlib import Path
 
+from video_processing.speed_engine import SpeedEngine
+
 # 延迟导入避免循环依赖：unified_render_service 定义 ResolvedClip / RenderLayer，
 # 本模块提供音频函数供 unified_render_service 调用。
 # 使用 from __future__ import annotations + TYPE_CHECKING 解决类型引用。
@@ -145,38 +147,104 @@ def concat_main_audio(
         # 单 clip，直接提取音频，截断到 min(clip有效时长, 视频总时长)
         clip = clips[0]
         effective_duration = clip_effective_duration(clip)
-        # 最终时长：取 clip 有效时长和视频总时长的较小值
-        # （视频总时长由主图层决定，但单 clip 场景下两者应该一致，仍做保护）
-        final_duration = effective_duration
+        speed = getattr(clip, "playback_speed", 1.0) or 1.0
+        if not isinstance(speed, (int, float)) or speed <= 0:
+            speed = 1.0
+
+        # 调速后时长
+        adjusted_duration = effective_duration / speed if abs(speed - 1.0) >= 1e-6 else effective_duration
+        # 最终时长：取调速后时长和视频总时长的较小值
+        final_duration = adjusted_duration
         if video_duration > 0 and (final_duration <= 0 or final_duration > video_duration):
             final_duration = video_duration
 
-        command = [
-            FFMPEG_BIN,
-            "-y",
-            "-i",
-            str(clip.local_path),
-            "-vn",
-            "-acodec",
-            "aac",
-            "-b:a",
-            "128k",
-        ]
-        if final_duration > 0:
-            command.extend(["-t", f"{final_duration:.3f}"])
-        command.append(str(output_path))
-        run_ffmpeg(command)
+        if abs(speed - 1.0) < 1e-6:
+            # 原速：简单命令行
+            command = [
+                FFMPEG_BIN,
+                "-y",
+                "-i",
+                str(clip.local_path),
+                "-vn",
+                "-acodec",
+                "aac",
+                "-b:a",
+                "128k",
+            ]
+            if final_duration > 0:
+                command.extend(["-t", f"{final_duration:.3f}"])
+            command.append(str(output_path))
+            run_ffmpeg(command)
+        else:
+            # 有调速：用 filter_complex + atempo
+            from video_processing.speed_engine import SpeedConfig
+            speed_engine = SpeedEngine()
+            config = SpeedConfig(speed=float(speed))
+            config.clamp()
+            atempo_filter = speed_engine.build_audio_filter(config)
+
+            filter_parts: list[str] = []
+            audio_filters = []
+            if effective_duration > 0:
+                audio_filters.append(f"atrim=0:{effective_duration:.3f}")
+                audio_filters.append("asetpts=PTS-STARTPTS")
+            if atempo_filter:
+                audio_filters.append(atempo_filter)
+
+            filter_parts.append(f"[0:a]{','.join(audio_filters)}[outa]")
+            if video_duration > 0 and final_duration < adjusted_duration:
+                filter_parts.append(f"[outa]atrim=0:{final_duration:.3f}[final_audio]")
+                final_label = "final_audio"
+            else:
+                final_label = "outa"
+
+            filter_complex = ";".join(filter_parts)
+            command = [
+                FFMPEG_BIN,
+                "-y",
+                "-i",
+                str(clip.local_path),
+                "-filter_complex",
+                filter_complex,
+                "-map",
+                f"[{final_label}]",
+                "-acodec",
+                "aac",
+                "-b:a",
+                "128k",
+                str(output_path),
+            ]
+            run_ffmpeg(command)
         return
 
     # 多 clip，用 filter_complex concat
     input_args: list[str] = []
     filter_parts: list[str] = []
+    speed_engine = SpeedEngine()
 
     for i, clip in enumerate(clips):
         input_args.extend(["-i", str(clip.local_path)])
         effective_duration = clip_effective_duration(clip)
+        speed = getattr(clip, "playback_speed", 1.0) or 1.0
+        if not isinstance(speed, (int, float)) or speed <= 0:
+            speed = 1.0
+
+        audio_filters: list[str] = []
         if effective_duration > 0:
-            filter_parts.append(f"[{i}:a]atrim=0:{effective_duration:.3f},asetpts=PTS-STARTPTS[a{i}]")
+            audio_filters.append(f"atrim=0:{effective_duration:.3f}")
+            audio_filters.append("asetpts=PTS-STARTPTS")
+
+        # 音频调速 — atempo 多级串联
+        if abs(speed - 1.0) >= 1e-6:
+            from video_processing.speed_engine import SpeedConfig
+            config = SpeedConfig(speed=float(speed))
+            config.clamp()
+            atempo_filter = speed_engine.build_audio_filter(config)
+            if atempo_filter:
+                audio_filters.append(atempo_filter)
+
+        if audio_filters:
+            filter_parts.append(f"[{i}:a]{','.join(audio_filters)}[a{i}]")
         else:
             filter_parts.append(f"[{i}:a]asetpts=PTS-STARTPTS[a{i}]")