Fix: Correct device placement for QuantizedLinear across all quantizers

Extends the previous fix for AWQ to GPTQ and GGUF quantizers.
Addresses an AttributeError where QuantizedLinear (an nn.Module)
was incorrectly passed to `move_to_device`, a function expecting
a tensor. This change ensures QuantizedLinear modules are moved to
their target device using the correct `.to(device)` method in
AWQ, GPTQ, and GGUF quantizers.

This commit ensures consistent and correct device handling for
quantized layers created by these methods.

Author: google-labs-jules[bot]

quantllm/quant/gguf.py

@@ -232,7 +232,7 @@ def _quantize_layer(
         chunk_size = 1024  # Adjust based on available memory
 
 
-        quantized = move_to_device(quantized, target_device)
+        quantized = quantized.to(target_device)
 
         # Copy bias if exists
         if layer.bias is not None:

quantllm/quant/gptq.py

@@ -202,7 +202,7 @@ def _quantize_layer(self, layer: nn.Linear, H: torch.Tensor) -> QuantizedLinear:
                 calibration="gptq"
             )
         )
-        quantized = move_to_device(quantized, target_device)
+        quantized = quantized.to(target_device)
 
         if layer.bias is not None:
             # layer is already on target_device