feat(voice): play TTS through tablet speakers via ElevenLabs proxy

Stage 2 of voice integration — centralizes TTS on the tablet so the
Python satellite no longer needs ElevenLabs credentials or mpv.

- app/api/voice/tts — POST {text, agent}, proxies to ElevenLabs
  streaming endpoint with flash_v2_5 default, returns audio/mpeg.
  Per-agent voice id via COSMO_TTS_VOICE / LUSYA_TTS_VOICE env.
- VoiceOverlay — on response/error events fetches TTS and plays via
  HTMLAudioElement; on wake event stops playback (barge-in). Dismiss
  timer extended by text length so long responses do not cut off.
- Autoplay caveat: browser may block first playback until user taps
  anywhere on the page (FKB: enable Force Autoplay to bypass).

app/api/voice/tts/route.ts

@@ -0,0 +1,75 @@
+export const dynamic = 'force-dynamic'
+export const runtime = 'nodejs'
+
+import { NextResponse } from 'next/server'
+
+const ELEVENLABS_BASE = 'https://api.elevenlabs.io/v1'
+const DEFAULT_MODEL = 'eleven_flash_v2_5'
+
+function getVoiceId(agent: string | undefined): string | null {
+  if (agent === 'lusya') return process.env.LUSYA_TTS_VOICE || null
+  return process.env.COSMO_TTS_VOICE || null
+}
+
+export async function POST(req: Request) {
+  const apiKey = process.env.ELEVENLABS_API_KEY
+  if (!apiKey) {
+    return NextResponse.json({ error: 'tts_not_configured' }, { status: 503 })
+  }
+
+  const body = await req.json().catch(() => null)
+  const text = typeof body?.text === 'string' ? body.text.trim() : ''
+  const agent = typeof body?.agent === 'string' ? body.agent : 'cosmo'
+
+  if (!text) {
+    return NextResponse.json({ error: 'text required' }, { status: 400 })
+  }
+  if (text.length > 4000) {
+    return NextResponse.json({ error: 'text too long (>4000)' }, { status: 400 })
+  }
+
+  const voiceId = getVoiceId(agent)
+  if (!voiceId) {
+    return NextResponse.json({ error: `no voice configured for agent=${agent}` }, { status: 503 })
+  }
+
+  const model = process.env.ELEVENLABS_MODEL || DEFAULT_MODEL
+
+  const upstream = await fetch(
+    `${ELEVENLABS_BASE}/text-to-speech/${encodeURIComponent(voiceId)}/stream?output_format=mp3_44100_64`,
+    {
+      method: 'POST',
+      headers: {
+        'xi-api-key': apiKey,
+        Accept: 'audio/mpeg',
+        'Content-Type': 'application/json',
+      },
+      body: JSON.stringify({
+        text,
+        model_id: model,
+        voice_settings: {
+          stability: 0.45,
+          similarity_boost: 0.75,
+          style: 0.25,
+          use_speaker_boost: true,
+        },
+      }),
+    }
+  )
+
+  if (!upstream.ok || !upstream.body) {
+    const errText = await upstream.text().catch(() => '')
+    return NextResponse.json(
+      { error: `elevenlabs_${upstream.status}`, detail: errText.slice(0, 300) },
+      { status: 502 }
+    )
+  }
+
+  return new Response(upstream.body, {
+    headers: {
+      'Content-Type': 'audio/mpeg',
+      'Cache-Control': 'no-store',
+      'X-Accel-Buffering': 'no',
+    },
+  })
+}