* implement composite full_like and simple unit test

* implement op tests for composite full_like op

* some modification as reviewers suggested
add cinn op test to CMakeLists.txt
fix code style

* fix code style

* modify input args of prim fill_any_like op

* resolve conflicts

* resolve conflicts

* modify python api and unit tests as suggested

* resolve conflicts

* resolve conflicts

* use framework.dtype to convert dtype in Op test

* add support of int64 add for xpu

* add transpose support for int64

* add randperm kernel

* fix randperm

* add distribute_fpn_proposal kernel

* fix comment

* add reduce_sum_int32

* [CodeStyle] fix flake8 cannot find the config on Windows

* empty commit; test=document_fix

* Add gru qat int8 test

* Change place of model downloading

* Update paddle/fluid/inference/tests/api/CMakeLists.txt
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* Correct flags names and add description

---------
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* [phi] move device_wrapper from fluid to phi

* [phi] fix ‘PADDLE_ENFORCE_XDNN_SUCCESS’ was not declared in this scope

* Added file

* Tests separated and rewritten, fixed fc_lstm_fuse_pass

* Resolve conflicts

* [Extension Operants] Extension supports tensor operants

* Polish fluid init_tensor_operants

* modify name

* merge develop

* original code

* build modify

* success 2*2

* fused dim=1 failed

* success

* modify static

* success for static except dim=1

* delete log

* tmp modify

* success

* success

* add fp1664

* delete fp16 cpu test

* stop windows test

* review modify

* modify tanh test

* modify tanh

* fix_conflixt

* modift static prim

* fix_conflict

* Update test_static_prim.cc

* update

* bug fix

* [C++ API GetAllocator] Add C++ GetAllocator interface

* move api to accurate directory

* main code

* unitest bug

* revert cmake

* add cumsum prim backward

* skip aixs=None test case

* fix op generante eror

* fix static test error

* remove unused code

* fix static test error

* skip cpu float16 test case

* skip eager cpu cumsum float16 test case

* add cinn test

* reshape flatten out

* Disable cinn single test

* remove cinn test

* reformat todo

* add prim in cumsum op test

* remove old test

* fix typro

* fix typro

* fix typro

* pass axis=None test case

* remove forward prim test

* remove same name axis

* modify install check: static graph parallel training to dynamic graph prallel training

* remove test code

* fix cyclic import

* fix typo

* fix fp16 bug of tensordot

* fix fp16 of tensordot

* fix fp16 of tensordot

* support fp16 on LocalResponseNorm

* add docs in avgpool3d

* add unittest for nn.DropOut2D

* add fp16

* add fp16 in docs of temporal_shift_op.cc

* Update test_dropout_op.py

* add silu composite rule

* fix code style.

* add silu fp16 unit test.

* forbid tensorrt_engine op's output is a persistable var

* Add flatten composite rule

* get the right xshape and pass func test

* add cinn unit test

* Remove cinn test, wait for it to be added after repair

* add comp test to test_flatten_contiguous_range_op.py

* remove func test on composite_ops

* Add comments to maybe_wrap_dim func

* remove commented code

* fix the problem with 0D tensor case

* add flatten split rule comment

* fix syntax issues

* block flatten on resnet_prim_cinn

* remove maybe_wrap_dim func

* Use none instead od xshape