* add assign op

* add test assign npu test

* dele if def
Co-authored-by: Noyjxer <1728722986@qq.com>

* accuracy op

* fix license

* fix

* add test and fix bug

* Support npu kernel for amp_check_finite_and_unscale_npu op

* support EnforceNotMet exception

* fix exception bug

* modify python unittest

* precommit

* update c++ unittest

* fix review

* fix review

* add reshape2 npu

* add reshpe2

* add relu npu

* fixed

* fix

* add matmulv2_npu

* add matmul

* add matmul

* add reduce any npu op

* add gather python unittest

* update c_plus unittest

* update python unittest

* del c++ unittest

* update c++ unittest

* update c++ unittest

* add fill_constant npu

* add fill_constant npu

* fix

* add scale npu

* fix

* fix

* add gather npu op

* code review done

* update python new line

* precommit

* fix review

* del commit

* add shape npu

* fix

* fix

* add npu kernel for equal op

* refine code

* add more ut

* update year

* init commit, add layer_norm npu kernel

* fix typo

* add unittest

* add unittest

* fix bug

* fix bug

* refine ut

* [NPU] Support npu op: (1) pow (2) pow_grad

* Support fp16

add ascend unittest

Ascendrc

* [NPU] Support executor with NPU

* Fix code according to reviews

* Fix code

* Add unittest for sub op npu

support parsing ascend rank table file

[Feature] Build parser to support distributed training

cleanup test_ascend_group.py

Add startup bash files of test_ascend_group

Add Hccl program group

Pass device_ids info from launch to trainer

Add distribution supported

* test=develop, fix fleet.metrics(mse, rmse, mae)

* test=develop, add distributed_infer

As the title

As the title

* optimize memcpy perf for kunlun

* remove useless unitest for kunlun mean

* minor