[ROCm][Bugfix][FP8] Make fp8 quant respect fused modules mapping (#16031)

Signed-off-by: mgoin <[email protected]>

Author: mgoin

vllm/model_executor/layers/quantization/fp8.py

@@ -116,7 +116,9 @@ def get_quant_method(self, layer: torch.nn.Module,
         from vllm.attention.layer import Attention  # Avoid circular import
 
         if isinstance(layer, LinearBase):
-            if is_layer_skipped(prefix, self.ignored_layers):
+            if is_layer_skipped(prefix=prefix,
+                                ignored_layers=self.ignored_layers,
+                                fused_mapping=self.packed_modules_mapping):
                 return UnquantizedLinearMethod()
             return Fp8LinearMethod(self)
         elif isinstance(layer, FusedMoE):