[SW-232491] Fix collective functions from vLLM (#250)

Author: kdamaszk

neural_compressor/torch/algorithms/fp8_quant/_core/vllm_functions.py

@@ -23,10 +23,10 @@
     tensor_model_parallel_all_gather = None
     tensor_model_parallel_all_reduce = None
 
-def get_vllm_row_parallel_collective_func():
+def get_vllm_column_parallel_collective_func():
     assert tensor_model_parallel_all_gather is not None, "Couldn't import vllm function tensor_model_parallel_all_gather"
     return tensor_model_parallel_all_gather
 
-def get_vllm_column_parallel_collective_func():
+def get_vllm_row_parallel_collective_func():
     assert tensor_model_parallel_all_reduce is not None, "Couldn't import vllm function tensor_model_parallel_all_reduce"
     return tensor_model_parallel_all_reduce