* [Auto Parallel] Add distributed mul for the old version

* transfer unsqueeze to phi

* fix conflict

* add squeeze

* add infershape

* fix xpu and npu error

* move activation

* fix bugs when run ce

* add full_batch_size_like phi kernel

* remove fill constant bs like

* update year

* [KP] fix compilation bug in phi

* delete the comment

* delete useless comment

* add kernel profiler

* add gpu timer tool

* remove warmup

* fix rocm complilation error

* Removed redundant use of declarations.h

* Fixed minor bug

* Fix test case timeout

* test=document_fix

* test=document_fix

* async prepare deps

* fix bug that std::future is not set

* add ut

* refine code

* fix standalone ut

* disable prof

* move mutable_data to context alloc

* move mutable_data to context alloc

* remvoe duplicate code

* 1. add the python api grad 2. add final and intermediate state vlog 3. change the python_api error logic

* add python api or close the check_eager=True

* fix the compatibility

* matmul

* disable unittests: test_elementwise_add_op test_scatter_nd_op test_gather_nd_op test_scatter_op test_index_sample_op test_elementwise_add_mkldnn_op

* move embeding to phi;

* update sig; test=develop

* move reset impl to phi; test=develop

* remove old register; test=develop

* fix cpu bf16 bug; test=develop

* fix lookup speed error

* polish code

* fix paddle throw type

* add reverse phi kernel

* add reverse infermeta

* remove original reverse op kernl & infershape

* fix context pool sync init

* add lock for insert

* optimize performance of  C++ API

* remove stop_data_transform flag temparily

* replace Backend by Place in C++ API

* fix left code

* fix test_to_api bug

* add out_size shape for graph_send_recv

* fix bug in register kernel: no const int& support

* add out_size in infermeta

* change unittest

* fix unittest

* fix out_size default value

* fix doc

* delete arg mapping

* add sig

* move -1 to 0

* move -1 to 0

* modify out and out_grad order in reduce_grad_kernel

* delete unsed boolReduceKernel

* fix conflict

* support inplace in eager_final state

* little change

* little bug

* fix group_norm address misalignment

* fix vectorize

* fix code

* fix vectorize length

* optimize code

* add sep line for conflict; test=document_fix

* revert

* add sep; test=document_fix

* Refine to_tensor for eager mode

* support gpu_pinned

* refine code

* support gpu_pinned copy_to

* fix layer.__setattr__

* support to_tensor for gpu_pinned

* fix unit test

* refine gpu_pinned

* restore the original code

* add is_gup_pinned() and refine eager.Tensor._copy_to()

* update readme test=document_fix

* update details

* update details,test=document_fix

* Support MaskedSelectGrad op with Kernel Primitive API

* remove duplicate code;

* add some line; test=document_fix

* switch eager mode and change it

* set default is eager

* set default is eager

* fix error; test=develop

* fix some error; test=develop

* update

* upd

* update code; test=develop

* update

* fix some bug; test=develop

* fix bug; test=develop

* fix bug; test=develop

* fix bug; test=develop

* fix error; test=develop

* format; test=develop
Co-authored-by: NJiabinYang <360788950@qq.com>