[P/D]Mooncake Layerwise Connector supports hybrid attention manager with multiple kvcache groups (#7022)

### What this PR does / why we need it?
Mooncake Layerwise Connector supports hybrid attention manager with
multiple kvcache groups.

### Does this PR introduce _any_ user-facing change?
Yes.

### How was this patch tested?
By CI.

- vLLM version: v0.16.0
- vLLM main:
15d76f74e2

---------

Signed-off-by: nwpu-zxr <zhouxuerong2@huawei.com>

vllm_ascend/utils.py

@@ -1073,7 +1073,11 @@ def refresh_block_size(vllm_config):
         return
 
     # TODO(MengqingCao): Remove the model_type check, after resolving the hidden error in get_kv_cache_groups.
-    if model_config.hf_text_config.model_type != "qwen3_next" and cache_config.block_size != 128:
+    if (
+        "qwen3_next" not in model_config.hf_text_config.model_type
+        and "qwen3_5" not in model_config.hf_text_config.model_type
+        and cache_config.block_size != 128
+    ):
         if cache_config.enable_prefix_caching or scheduler_config.enable_chunked_prefill:
             logger.info("Block size is set to 128 if prefix cache or chunked prefill is enabled.")
             cache_config.block_size = 128