* refine bf16 amp-o1 logic

* refine amp GLOG

* refine unittest

* refine unittest

* remove empty kernel in fluid and adjust the param of empty dev_api

* polish code

* revert fluid empty kernel

* fix selected_rows bug in C++ API

* add optional for C++ APIO

* data transform support optional

* remove data transform for optional vector<Tensor>

* adjust some format of funtcion

* fix empyt bug

* add new profiler components

* fix bug

* [KP] Unify .cu and .xpu files with .kps files

* fix CI bug in GPU and modify the list

* fix conflict

* modify the date

* [Phi] Add ClearHolder when re-alloc on new place in DeviceContext

* fix hostAlloc

* foix inferRT unittest

* remove dev_ctx ptr

* fix pylayer problem with amp

* add ut

* refine code

* remove SetAllocationForOutputTenosr

* add place param for copy kernel

* recover SetAllocationForOutputTenosr

* polish code

* fix empty_dev api bug

* test=allcases

* test=allcases

* fix bug

* recover empty

* recover modify

* Move GumbelSoftmax OP to phi

* platform::errors -> phi::errors; GumbelSoftmaxGradInferMeta -> backend.h/cc

* Use axis util in kernel impl

* Remove namespace platform::errors

* Use GetCPUEngine in Device Context

* Support custom implement for C++ API

* rename api_invoke_impl to api_custom_impl

* remove manual_api

* delete mutable_data in copy_to api

* fix problem of copy_to

* add unittest for infer_meta_fn_factory

* fix split cofig in yaml

* fix split cofig in yaml

* modify sum api yaml

* add copy_to wrapped infermeta

* rollback copy impl

* Move BilinearTensorProduct OP to phi

* Set dtype for Infermeta

* Support Eager Hook, expose interface to python

* Fix CI issue

* add multi tensor apply l2 norm

* add multi_tensor_apply code

* make sizeof(TensorMeta) smalller

* move code to distributed_fused_lamb_op.cu

* remove useless FLAGS

* move eye OP to pten

* move size OP to pten

* merge develop

* fix merge

* move files

* move erfinv OP to phi

* remove comment

* move pixel_shuffle OP to phi

* remove comment

* fix PT_REGISTER

* fix NPU

* fix CR

* remove size_sig.cc for PR-CI-Coverage

* disable some distribute test case when in CPU test env

* disable some test case when in CPU test env

* fix

* [phi]migrate increment addmm multinomial cholesky InferShapes to phi

* set_dtype and mod MultinomialFunctor

* [ROCm] fix Managed Memory Alloc on HIP, test=develop

* update, test=develop

* Optimize perf of softmax_with_cross_entropy

* fix

* fix

* fix accuracy error

* add ele_add

* add ele_mul

* add ele_sub

* sovle conflict

* fix npu

* refine ele_add

* add ele_mul unittest

* refine ele_sub

* refine ci

* refine unittest

[Phi] mv kernel 

* refine randint kernel

* refine randperm kernel

* refine unbind kernel

* support op seed

* support cudnn kernel moving

* polish cmake rules

* add unittest for coverage

* remove orig kernel

* remove softmax cudnn kernel

* fix softmax test failed

* fix npu func error

* resolve conflict

* rename gpu dnn kernels

* fix name rule error

* fix compile error

* update fp16 namespace

* fix bugs

* fix bugs

* Fix compile error on cuda_arch less than 700.