[Edge][WP2] Add initialization timeline and cold-start profiler

* vllm_omni/engine/init_timeline.py: InitTimeline phase collector; engine
  phases (stages_init, g3_device_lock, g2_spawn/ready) and worker phases
  (proc_entry, engine_core_init, init_device, load_model/load_weights,
  talker_mtp_init, attn_prewarm, cudagraph_capture, talker_mtp_capture,
  code2wav_cudagraph, predictor_setup_compile/warmup/cudagraph) recorded via
  worker_phase/timed_phase; workers append JSONL under
  VLLM_OMNI_INIT_TIMELINE_DIR (inherited at spawn) and the engine merges them.
  Enabled by VLLM_OMNI_INIT_TIMELINE=<json> or log_stats; table logged at INFO,
  partial table on orchestrator startup timeout, suggested timeouts logged.
* omni_base.py: named deploy profiles default to stage_init_timeout=900 /
  init_timeout=1500 unless given explicitly.
* gpu_model_runner._init_talker_mtp tolerates max_cudagraph_capture_size=None;
  gpu_ar_model_runner.shutdown skips CUDA graph teardown on CPU.
* benchmarks/tts/init_profile.py: runs stream_latency_bench --init-only per
  configuration (default, cold_cache, eager_stage1, eager_both,
  parallel_stage_init, edge, ...) and tabulates phase durations.
* stream_latency_bench: --parallel-stage-init / --deploy-profile flags; the
  resource poller no longer shadows Thread._stop and skips GPU sampling on CPU.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

vllm_omni/worker/gpu_generation_worker.py

@@ -11,6 +11,7 @@ from vllm.v1.utils import report_usage_stats
 from vllm.v1.worker.gpu_worker import init_worker_distributed_environment
 from vllm.v1.worker.workspace import init_workspace_manager
 
+from vllm_omni.engine.init_timeline import timed_phase
 from vllm_omni.platforms import current_omni_platform
 from vllm_omni.worker.base import OmniGPUWorkerBase
 from vllm_omni.worker.gpu_generation_model_runner import GPUGenerationModelRunner
@@ -30,6 +31,7 @@ class GPUGenerationWorker(OmniWorkerMixin, OmniGPUWorkerBase):
     model_runner_cls = GPUGenerationModelRunner
 
     @instrument(span_name="Init device")
+    @timed_phase("init_device")
     def init_device(self):
         if self.device_config.device_type in ("cuda", "musa"):
             # This env var set by Ray causes exceptions with graph building.