paddle.scatter_nd_add

* remove apis in fluid.ops

* fix test_activation_nn_grad

* fix circle import error

* fix ops

* fix cos

* fix divide not inplace

* remove lazy-import part

remove crop which is not used in Paddle 2.0

* remove relu6 test case under fluid

* fix relu6 test case in mkldnn_elt_act_fuse_pass

replace paddle.fluid.layers.selu with paddle.nn.functional.selu

* Remove API: gather
	replace the paddle.fluid.layers.gather with paddle.gather

* modify the call of gather from old style to new style

* remove range

* [CustomPlace] fix amp

* [CustomPlace] fix amp

* fix ut because of too long time matmul fp16

remove pad_constant_like which is not used in paddle 2.0

* [AutoParallel] selective recompute

* add cmakelist

* clip extra and intermediate output of op

* fix bug

* fix bug

* polich code

* polich log

* remove unstack in nn.py under fluid

* remove unstack under fluid

* clean fluid elementwise_pow, remove API

* clean elem_pow doc

* clean elementwise_mod

* clean elementwise min, floordiv, mod

* [NPU] add _npu_identity op and api, test=develop

* fix doc

* address comments

* remove swish in nn.py under fluid

* fix tswish test case

* int scale

* round

* revert commit

* standard api

* fix xpu bugs

* add_cast_bool

* cast

* add vectorized bfloat16 atomicAdd

* fix compile error

* fix compile error again

* fix V100 compile error

* fix V100 compile again

* clean fluid elementwise_min

* fix elementwise_min op testcase