fix: ruff import sort order in test_pytorch.py

Signed-off-by: 김대익 <33992354+dik654@users.noreply.github.com>

Author: dik654

src/llmcompressor/modifiers/README.md

@@ -37,7 +37,7 @@ bin using a scale and (optional) zero point. This basic quantization algorithm i
 suitable for FP8 quantization. A variety of quantization schemes are supported via the 
 [compressed-tensors](https://github.com/neuralmagic/compressed-tensors) library. 
 
-### [GPTQ](./quantization/gptq/base.py)
+### [GPTQ](./gptq/base.py)
 One-shot algorithm that uses calibration data to select the ideal bin for weight quantization. 
 This algorithm is applied on top of the basic quantization algorithm, and affects weights only.
 The implementation is based on [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://arxiv.org/pdf/2210.17323). The algorithm is very similar to SparseGPT: A small amount of calibration data is used 

src/llmcompressor/pipelines/sequential/README.md

@@ -1,4 +1,4 @@
 # Sequential Pipeline #
 The sequential pipeline is a data pipeline, primarily used for compressing models with the
-[GPTQModifier](/src/llmcompressor/modifiers/quantization/gptq/base.py) or the
+[GPTQModifier](/src/llmcompressor/modifiers/gptq/base.py) or the
 [SparseGPTModifier](/src/llmcompressor/modifiers/pruning/sparsegpt/base.py).

tests/llmcompressor/pytorch/modifiers/pruning/sparsegpt/test_pytorch.py

@@ -1,8 +1,8 @@
 import pytest
 import torch
 
-from llmcompressor.modifiers.pruning.sparsegpt import SparseGPTModifier
 from llmcompressor.modifiers.gptq import GPTQModifier
+from llmcompressor.modifiers.pruning.sparsegpt import SparseGPTModifier
 from tests.llmcompressor.modifiers.conf import LifecyleTestingHarness
 from tests.llmcompressor.pytorch.helpers import LinearNet