### What this PR does / why we need it?
Add Qwen3Next support in main
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.11.2
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.2
---------
Signed-off-by: SunnyLee219 <3294305115@qq.com>