[MR V2] Migrate to Model Runner V2

Squashed migration rebased onto current main. Collapses the 46-commit
dev/migrate-MR-v2 history (10 merge commits + superseded pre-rebase work)
into a single commit, so the diff reflects only the genuine MR-V2 feature
(worker_v2/ runners + model_states, omni schedulers, stage engine pipeline,
pooling_output serialization bridge). Content already merged to main
separately (step_audio2, tts_adapters, npu/_310p, etc.) drops out.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tests/examples/online_serving/test_qwen2_5_omni.py

@@ -115,13 +115,16 @@ def test_send_multimodal_request_002(omni_server) -> None:
 @hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards={"cuda": 4, "rocm": 2})
 @pytest.mark.parametrize("omni_server", test_params, indirect=True)
 def test_send_multimodal_request_003(omni_server) -> None:
+    env = os.environ.copy()
+    env["MODEL"] = omni_server.model
+
     command = [
         "bash",
         os.path.join(example_dir, "qwen2_5_omni/run_curl_multimodal_generation.sh"),
         "mixed_modalities",
     ]
 
-    result = run_cmd(command)
+    result = run_cmd(command, env=env)
 
     text_content = extract_content_after_keyword("Output of request:", result)