Merge remote-tracking branch 'origin/main' into qwen-image-2.1

# Conflicts:
#	vllm_omni/diffusion/executor/multiproc_executor.py
#	vllm_omni/diffusion/offloader/sequential_backend.py

vllm_omni/diffusion/offloader/sequential_backend.py

@@ -13,7 +13,6 @@ from vllm_omni.diffusion.hooks import HookRegistry, ModelHook
 from vllm_omni.platforms import current_omni_platform
 
 from .base import OffloadBackend, OffloadConfig, SupportsModelCpuOffload
-from .config import DIT_COMPONENT
 from .module_residency import PinnedModuleStager
 from .plan_resolver import resolve_offload_plan
 
@@ -385,6 +384,7 @@ class ModelLevelOffloadBackend(OffloadBackend):
         encoders = [component.module for component in resolved.encoders]
         vaes = [component.module for component in resolved.vaes]
         residents = [component.module for component in resolved.residents]
+        selected_dits = [component.module for component in resolved.dits if component.selected]
         selected_encoders = [component.module for component in resolved.encoders if component.selected]
 
         all_modules = [*dits, *encoders, *vaes, *residents]
@@ -412,7 +412,7 @@ class ModelLevelOffloadBackend(OffloadBackend):
                 device=self.device,
                 pin_memory=self.config.pin_cpu_memory,
                 use_hsdp=self.config.use_hsdp,
-                offload_dit_modules=(dits if self.config.offloads(DIT_COMPONENT) else ()),
+                offload_dit_modules=selected_dits,
                 offload_encoder_modules=selected_encoders,
                 # Fixed DiT staging keeps decode-graph weight pointers valid;
                 # other platforms keep plain move semantics.