[Diffusion] Add native paged KV backend for diffusion workers (#6102)

Signed-off-by: Acerak01-fy <wfy2003324@163.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>

vllm_omni/diffusion/diffusion_kv/config.py

@@ -37,3 +37,8 @@ def parse_diffusion_kv_cache_mode(value: object) -> DiffusionKVCacheMode:
             "use 'dense_legacy' or 'paged_scheduler'"
         )
     return mode
+
+
+def is_scheduler_paged_kv_mode(mode: DiffusionKVCacheMode) -> bool:
+    """Return whether a parsed cache mode uses Scheduler-owned paging."""
+    return mode is DiffusionKVCacheMode.PAGED_SCHEDULER