[CI] remove old quantization model (vllm-project#1003)

remove old quantization model, and new models will be added to testcase
later.

Signed-off-by: 22dimensions <waitingwind@foxmail.com>
Signed-off-by: wangxiaoxin (A) <wangxiaoxin7@huawei.com>

Author: 22dimensions

tests/singlecard/test_offline_inference.py

@@ -37,6 +37,10 @@
     "Qwen/Qwen3-0.6B-Base",
 ]
 MULTIMODALITY_MODELS = ["Qwen/Qwen2.5-VL-3B-Instruct"]
+
+QUANTIZATION_MODELS = [
+    "vllm-ascend/Qwen2.5-0.5B-Instruct-W8A8-new",
+]
 os.environ["PYTORCH_NPU_ALLOC_CONF"] = "max_split_size_mb:256"
 
 QUANTIZATION_MODELS = [