rtx-csm: Phase 6f.warmup — pre-warm Generator at server boot

Issues one throwaway Generator.generate() call after model load + before
accepting connections. Pays Metal kernel compilation + first-frame KV
cache init up-front so the first real user turn doesn't carry that
overhead.

A/B (mock LLM, FP CSM, M-series Metal, 10.43s LibriSpeech in):
  no warm-up  sentence[0] tts_gen=5679ms  total=24818ms
  with warm-up sentence[0] tts_gen=4272ms total=18654ms

Warm-up cost at boot: 1417ms. Direct saving on first-sentence TTS gen:
~1.4s. The warm-up pays for itself on the first turn and is amortized
to zero across the server's lifetime.

Failures during warm-up are logged at warn level and the server boots
anyway — first turn falls back to the original cold-start behavior.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>

crates/models/rtx-csm/examples/converse_server.rs

@@ -441,6 +441,33 @@ async fn main() -> Result<()> {
     };
     tracing::info!("models loaded");
 
+    // Warm up the Generator: do one throwaway generate() so Metal kernel
+    // compilation + first-frame KV cache init happen now, not on the
+    // first user turn. Empirically this eliminates ~5-6s of dead time
+    // from the first-sentence TTS gen latency. Cost is paid once at
+    // boot. The throwaway audio is discarded.
+    let warmup_t = std::time::Instant::now();
+    {
+        let opts = GenerateOptions {
+            max_audio_ms: 200,
+            seed: 0,
+            ..GenerateOptions::default()
+        };
+        match generator.generate("Hi.", cli.speaker, &[], opts) {
+            Ok(_) => {
+                tracing::info!(
+                    "generator warm-up complete ({}ms)",
+                    warmup_t.elapsed().as_millis()
+                );
+            }
+            Err(e) => {
+                tracing::warn!(
+                    "generator warm-up failed (non-fatal, first turn will be slower): {e}"
+                );
+            }
+        }
+    }
+
     let auth_token = cli
         .auth_token
         .or_else(|| std::env::var("RTX_AUTH_TOKEN").ok());