[Bugfix] Preserve default diffusion sampling params in default stage (#2780)

Signed-off-by: David Chen <530634352@qq.com>

tests/entrypoints/openai_api/test_image_server.py

@@ -177,7 +177,7 @@ def test_client(mock_async_diffusion):
         [BaseModelPath(name="Qwen/Qwen-Image", model_path="Qwen/Qwen-Image")]
     )
     app.state.args = Namespace(
-        default_sampling_params='{"0": {"num_inference_steps":4, "guidance_scale":7.5}}',
+        default_sampling_params='{"0": {"num_inference_steps":4, "guidance_scale":7.5, "generator_device":"cpu"}}',
         max_generated_image_size=1024 * 1792,
     )
 
@@ -200,7 +200,7 @@ def async_omni_test_client():
         SimpleNamespace(stage_type="diffusion"),
     ]
     app.state.args = Namespace(
-        default_sampling_params='{"1": {"num_inference_steps":4, "guidance_scale":7.5}}',
+        default_sampling_params='{"1": {"num_inference_steps":4, "guidance_scale":7.5, "generator_device":"cpu"}}',
         max_generated_image_size=1048576,  # 1024*1024 to support resolution tests
     )
     return TestClient(app)
@@ -222,7 +222,7 @@ def async_omni_rgba_test_client():
         SimpleNamespace(stage_type="diffusion"),
     ]
     app.state.args = Namespace(
-        default_sampling_params='{"1": {"num_inference_steps":4, "guidance_scale":7.5}}',
+        default_sampling_params='{"1": {"num_inference_steps":4, "guidance_scale":7.5, "generator_device":"cpu"}}',
         max_generated_image_size=1048576,
     )
     return TestClient(app)
@@ -244,7 +244,7 @@ def async_omni_stage_configs_only_client():
     # Intentionally do not populate app.state.stage_configs. Refactored
     # AsyncOmni exposes stage_configs on the engine instance.
     app.state.args = Namespace(
-        default_sampling_params='{"1": {"num_inference_steps":4, "guidance_scale":7.5}}',
+        default_sampling_params='{"1": {"num_inference_steps":4, "guidance_scale":7.5, "generator_device":"cpu"}}',
         max_generated_image_size=1024 * 1792,
     )
     return TestClient(app)
@@ -960,6 +960,7 @@ def test_image_edit_parameter_default(async_omni_test_client):
     assert captured_sampling_params.num_outputs_per_prompt == 1
     assert captured_sampling_params.num_inference_steps == 4
     assert captured_sampling_params.guidance_scale == 7.5
+    assert captured_sampling_params.generator_device == "cpu"
 
     # Test that a size exceeding max_generated_image_size returns 400
     response = async_omni_test_client.post(
@@ -993,6 +994,7 @@ def test_image_edit_parameter_default_single_stage(test_client):
     assert captured_sampling_params.num_outputs_per_prompt == 1
     assert captured_sampling_params.num_inference_steps == 4
     assert captured_sampling_params.guidance_scale == 7.5
+    assert captured_sampling_params.generator_device == "cpu"
 
     # Size exceeding max_generated_image_size (1024*1792) returns 400
     response = test_client.post(