rebase: adapt to vLLM 0.29 callback, embedding and video decoder changes

Four unrelated 0.29 surface changes, each reproduced locally against the
pinned f5c3cc24 build:

- PromptUpdateDetails.is_embed now takes only the full token list. The
  Bagel img2img replacement still declared (tok, seq), which crashed
  stage-0 engine core startup and produced the Distributed and Qwen3-Omni
  "no live replica" failures rather than anything KV related.

- dispatch_unquantized_gemm gained a linear_backend argument and is called
  with it. The CPU-forcing test stubs took no arguments; they now accept
  and ignore whatever upstream passes so a further signature change does
  not break them again.

- VocabParallelEmbedding gained a fused kernel that engages whenever
  tp_size > 1 on CUDA, and it has no CPU implementation. The encoder TP
  fixtures drive CPU tensors under a faked TP group, so they pin the
  portable masked path.

- The pyav video decoder backend was removed; valid options are now
  opencv, torchcodec, pynvvideocodec and deepstream. The OpenAI video
  endpoint hardcoded pyav, which turned every video request into a 400.
  It moves to opencv, the upstream default.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0146Xu5Sgj4vXxLxvLjV5YPj
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>

tests/diffusion/models/t5_encoder/test_t5_encoder_tp.py

@@ -54,6 +54,13 @@ def setup_tp_group(monkeypatch, mocker):
         lambda: 0,
     )
 
+    # vLLM 0.29 added a CUDA-only fused embedding kernel that engages whenever
+    # tp_size > 1; these tests drive CPU tensors, so keep the portable masked path.
+    monkeypatch.setattr(
+        "vllm.model_executor.layers.vocab_parallel_embedding.current_platform.is_cuda",
+        lambda: False,
+    )
+
     # TP group
     mock_tp_group = mocker.MagicMock()
     mock_tp_group.world_size = 2