[Feat] DiffusionEngine Support async batch infer (#2729)

Signed-off-by: Semmer <semmer@live.cn>
Signed-off-by: jader <yjader@foxmail.com>
Signed-off-by: asukaqaq-s <1311722138@qq.com>
Co-authored-by: asukaqaq-s <1311722138@qq.com>
Co-authored-by: jader <yjader@foxmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>

vllm_omni/engine/async_omni_engine.py

@@ -1329,7 +1329,7 @@ class AsyncOmniEngine:
         final_output_type = "audio" if model_class_name and supports_audio_output(model_class_name) else "image"
 
         stage_engine_args = {
-            "max_num_seqs": 1,
+            "max_num_seqs": kwargs.get("max_num_seqs") or 1,
             "parallel_config": parallel_config,
             "model_class_name": kwargs.get("model_class_name", None),
             "step_execution": kwargs.get("step_execution", False),