improve the robustness of parsing vlms config in AutoRound (#18894)

Signed-off-by: wenhuach21 <wenhua.cheng@intel.com>

vllm/model_executor/layers/quantization/auto_round.py

@@ -116,7 +116,8 @@ class AutoRoundConfig(QuantizationConfig):
 
         quantized = True
         if self.block_name_to_quantize:
-            quantized = any(name in layer_name
+            quantized = any(
+                layer_name.startswith(name)
                 for name in self.block_name_to_quantize)
         elif isinstance(layer, ParallelLMHead):
             quantized = False