[Model] Add user-configurable task for models that support both generation and embedding (#9424)

tests/entrypoints/openai/test_serving_chat.py

@@ -22,12 +22,12 @@ class MockHFConfig:
 
 @dataclass
 class MockModelConfig:
+    task = "generate"
     tokenizer = MODEL_NAME
     trust_remote_code = False
     tokenizer_mode = "auto"
     max_model_len = 100
     tokenizer_revision = None
-    embedding_mode = False
     multimodal_config = MultiModalConfig()
     hf_config = MockHFConfig()
 

tests/entrypoints/openai/test_vision.py

@@ -23,6 +23,8 @@ TEST_IMAGE_URLS = [
 @pytest.fixture(scope="module")
 def server():
     args = [
+        "--task",
+        "generate",
         "--dtype",
         "bfloat16",
         "--max-model-len",