feat: Runware FLUX.2 image + lazy per-beat TTS (#5)

Reduce median scene-load latency from ~30-80s to ~17-25s by switching image generation to Runware FLUX.2 [klein] 9B KV and moving per-beat TTS synthesis off the scene response into a new lazy /api/beat-audio endpoint with hard timeout + abort support.

- feat(image): migrate to Runware FLUX.2 [klein] 9B KV — task-array API, $0.001/image, sub-second inference.
- feat(tts): split /api/scene into directScene + image + voicedesign-provisioning; lazily synth per beat via /api/beat-audio with 15s hard timeout + AbortSignal threaded to MiMo so timed-out calls don't keep burning sockets/quota; client fans out per-beat fetches on scene-id change with abort + identity-check finally to prevent cross-scene beat-id collisions.
- refactor(tts): slim BeatAudioRequest to { beat, voice } — ~800KB per-beat upload dropped to ~160KB by sending only the speaker's voice instead of the full session.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

apps/web/app/api/beat-audio/route.ts

@@ -0,0 +1,36 @@
+import { requestBeatAudio } from "@yume/engine";
+import type { BeatAudioRequest } from "@yume/types";
+import { NextResponse } from "next/server";
+import { loadEngineConfig } from "@/lib/config";
+
+export const runtime = "nodejs";
+// The synth itself has a 15s per-call ceiling in the engine. 30s here just
+// covers JSON parsing + outbound network buffer.
+export const maxDuration = 30;
+
+export async function POST(req: Request) {
+  let body: BeatAudioRequest;
+  try {
+    body = (await req.json()) as BeatAudioRequest;
+  } catch {
+    return NextResponse.json({ error: "Invalid JSON" }, { status: 400 });
+  }
+
+  if (!body.beat?.id || !body.beat?.line || !body.voice?.referenceAudioBase64) {
+    return NextResponse.json(
+      { error: "beat.id, beat.line and voice.referenceAudioBase64 are required" },
+      { status: 400 },
+    );
+  }
+
+  try {
+    const config = loadEngineConfig();
+    const result = await requestBeatAudio(config, body);
+    return NextResponse.json(result);
+  } catch (err) {
+    // Engine already swallows synth errors and returns audio:null. Anything
+    // that reaches here is config-level — surface so the client can log it.
+    const message = err instanceof Error ? err.message : "Unknown error";
+    return NextResponse.json({ error: message }, { status: 500 });
+  }
+}