[MM][Model] Remove Qwen2-VL modeling files (#4534)

### What this PR does / why we need it?

Following #4349, remove
Qwen2-VL modeling files.


- vLLM version: v0.11.2
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.2

---------

Signed-off-by: shen-shanshan <[email protected]>

Author: shen-shanshan

tests/ut/models/test_qwen2_vl.py

@@ -2,10 +2,6 @@
 
 
 def register_model():
-    ModelRegistry.register_model(
-        "Qwen2VLForConditionalGeneration",
-        "vllm_ascend.models.qwen2_vl:AscendQwen2VLForConditionalGeneration")
-
     ModelRegistry.register_model(
         "Qwen3VLMoeForConditionalGeneration",
         "vllm_ascend.models.qwen3_vl:AscendQwen3VLMoeForConditionalGeneration")