* bugfix (#45332)

* dist embedding support lookup table v1

* add unitest

* customize wait_comm

* group gradients

* bugfix

* update program

* adapt gradient merge

* fix op_role

* fix strategy

* prune d2h memcpy for fp16 pass

* use yapf to format all python file

* yapf exclude two unittests file for they rely on writing and reading file, and format will break them

* disable diff_py_file because too many diff files cause command following failed

* use original id in dist_op_context.grad_op_id_to_op_id

* del assert

* remove redundant map

* adapot dist op

* [Auto Parallel] Support the auto completion of while_op

* add dist_fill_constant_batch_size_like

* align infer  accuracy

*  add FP16 Pass 

* Support the auto completion of while_op

*  acc aligned