[diffusion] CI: add test for cache-dit (#16204)

Co-authored-by: Mick <mickjagger19@icloud.com>

python/sglang/multimodal_gen/test/server/testcase_configs.py

@@ -153,6 +153,7 @@ class DiffusionServerArgs:
     lora_path: str | None = None  # LoRA adapter path (HF repo or local path)
 
     dit_layerwise_offload: bool = False
+    enable_cache_dit: bool = False
 
 
 @dataclass(frozen=True)
@@ -304,6 +305,17 @@ ONE_GPU_CASES_A: list[DiffusionTestCase] = [
         ),
         T2I_sampling_params,
     ),
+    DiffusionTestCase(
+        "qwen_image_t2i_cache_dit_enabled",
+        DiffusionServerArgs(
+            model_path="Qwen/Qwen-Image",
+            modality="image",
+            warmup_text=1,
+            warmup_edit=0,
+            enable_cache_dit=True,
+        ),
+        T2I_sampling_params,
+    ),
     DiffusionTestCase(
         "flux_image_t2i",
         DiffusionServerArgs(
@@ -414,14 +426,14 @@ ONE_GPU_CASES_B: list[DiffusionTestCase] = [
     ),
     # NOTE(mick): flaky
     # DiffusionTestCase(
-    #     id="hunyuan_video",
-    #     model_path="hunyuanvideo-community/HunyuanVideo",
-    #     modality="video",
-    #     prompt="A curious raccoon",
-    #     output_size="720x480",
-    #     warmup_text=0,
-    #     warmup_edit=0,
-    #     custom_validator="video",
+    #     "hunyuan_video",
+    #     DiffusionServerArgs(
+    #         model_path="hunyuanvideo-community/HunyuanVideo",
+    #         modality="video",
+    #     ),
+    #     DiffusionSamplingParams(
+    #         prompt=T2V_PROMPT,
+    #     ),
     # ),
     DiffusionTestCase(
         "flux_2_ti2i",