fix: require manual subject reconstruction generation

web/components/ad-recreation-board.tsx

@@ -3318,8 +3318,7 @@ function SourceSubjectPipeline({
     if (current.length >= RECONSTRUCTION_FRAME_LIMIT) {
       toast.warning(`${reconstructionModeConfig(mode).label}最多保留 ${RECONSTRUCTION_FRAME_LIMIT} 张参考帧，已替换为最近拖入的组合。`)
     }
-    toast.info(`已加入${reconstructionModeConfig(mode).label}：${frame.timestamp.toFixed(1)}s，开始生成 6 视图。`)
-    void generateSubjectPack(mode, next)
+    toast.info(`已加入${reconstructionModeConfig(mode).label}：${frame.timestamp.toFixed(1)}s。放好参考后点击生成 6 视图。`)
   }
 
   const removeConversionFrame = (mode: SubjectReconstructionMode, frameIndex: number) => {
@@ -3488,7 +3487,7 @@ function SourceSubjectPipeline({
         </div>
         <div className="max-h-[520px] min-h-[410px] overflow-y-auto rounded-md border border-white/10 bg-black/32 p-2 2xl:max-h-[600px] 2xl:min-h-[500px]">
           <div className="mb-2 rounded-md border border-[#d6b36a]/18 bg-[#d6b36a]/[0.06] px-2.5 py-2 text-[10px] leading-snug text-white/62">
-            拖入 1-3 张参考帧到对应方向；系统只做参考重构，不复制原人、原脸或原画面。
+            先拖入 1-3 张参考帧到对应方向，放好后再点击生成；系统只做参考重构，不复制原人、原脸或原画面。
           </div>
           <div className="space-y-2">
             {RECONSTRUCTION_MODES.map((modeConfig) => {