[SW-224836] disable test_mixed_precision_gptq_fp8_quant_only_nlp (#208)

Co-authored-by: linoy buchnik <[email protected]>

Author: ulivne

test/3x/torch/quantization/fp8_quant/test_gptq_mixed_precision.py

@@ -63,6 +63,7 @@ def teardown_class(self):
     Calibrates the w4a16 model on N num of samples, and compares the same samples with the hybrid model.
     """
     @torch.no_grad()
+    @pytest.mark.skip(reason="This test is temporarily disabled SW-224890")
     def test_mixed_precision_gptq_fp8_quant_only_nlp(self):
         from neural_compressor.common import set_random_seed
         set_random_seed(12345)