feat: per-call LLM timing, speaker ID, cancel captures on leave

- llm.py: log each Ollama call's caller + total/load/prompt/gen durations
  so a slow voice turn is attributable to a specific internal call
  (router/enrichment/digest/main); a RELOAD marker flags cold reloads.
- voice.ts: track in-flight Opus captures and abort them on session
  destroy(); drop any utterance that finishes after the user left, so no
  trailing post-leave VAD turns are reported.
- userbot.ts: show the speaker's Discord user ID on each transcript line
  (answered and dropped) so it's clear whose audio produced the turn.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

bot/src/userbot.test.ts

@@ -53,6 +53,14 @@ test("formatTurnMessage reports a dropped turn with only the listening stage", (
   expect(msg).not.toContain("🔊 TTS");
 });
 
+test("formatTurnMessage shows the speaker user ID when provided", () => {
+  const answered = formatTurnMessage({ user: "12345", transcript: "안녕", reply: "하이" });
+  expect(answered).toContain("👤 12345");
+  const dropped = formatTurnMessage({ user: "67890", transcript: "", reply: "", note: "음성 아님(VAD 차단)" });
+  expect(dropped).toContain("👤 67890");
+  expect(dropped).toContain("❌ 음성 아님(VAD 차단)");
+});
+
 test("formatTurnMessage falls back to the plain line when no timing is present", () => {
   const msg = formatTurnMessage({ transcript: "안녕", reply: "하이" });
   expect(msg).toBe('🎤 들음 → 🗣️ "안녕"\n🤖 답변: 하이');