[BugFix][CI] Address review follow-ups from v0.24.0 rebase (#4709)

Addresses Gaohan123's post-approval review comments on #4709:

- pipeline_dreamzero: restore platform-dispatched synchronize() instead of
  torch.cuda so DreamZero keeps working on non-CUDA hardware (regression
  of #4399 introduced during the rebase).
- omni_scheduler_mixin: drop the VLLM_OMNI_INPUT_WAIT_TIMEOUT_S env-var
  override; the full-payload input-wait timeout is now a plain constant.
- diffusion_worker: implement _DiffusionVllmModelConfig.is_diffusion by
  mirroring upstream ModelConfig.is_diffusion (dLLM canvas_length check)
  and document why omni DiT pipelines must not enable dLLM paths.
- qwen3_tts: make _embedding_dtype a class attribute on the talker and
  prompt-embeds builder so __new__-based test stubs inherit it; remove
  the (duplicated) manual assignments from the preprocess test.
- .buildkite: document why the rebase raised per-step timeouts (steps
  were killed by the old limits, masking real failures; e.g. CustomVoice
  passes in ~9-15 min but cold runs were killed at 20m).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>

vllm_omni/diffusion/worker/diffusion_worker.py

@@ -83,7 +83,13 @@ class _DiffusionVllmModelConfig:
 
     @property
     def is_diffusion(self) -> bool:
-        return False
+        # Duck-types ``vllm.config.ModelConfig.is_diffusion``, which detects
+        # *discrete diffusion (dLLM) text models* via ``hf_config.canvas_length``
+        # and gates dLLM-only scheduler/spec-decode paths in vLLM.  Omni's DiT
+        # pipelines run in their own engine and must never enable those paths,
+        # so mirror the upstream check (False for every DiT model) instead of
+        # leaving the attribute missing.
+        return getattr(self.hf_config, "canvas_length", None) is not None
 
 
 def _make_diffusion_vllm_model_config(od_config: OmniDiffusionConfig) -> _DiffusionVllmModelConfig: