* add mul and matmul quantization, test=develop
* add test for matmul, test=develop

* DataParallel state dict don't include _layers.; test=develop

* add unitest of data parallel; test=develop

* add load state test; test=develop

* polish grad op check, test=develop, test=document_fix

* keep op_use_default_grad_maker.spec to avoid conflict, test=develop, test=document_fix

* remove _optimized_guard in dygrahpe_mode test=develop

* remove comment code test=develop

* remove list append test=develop

* remove list append test=develop

* polish regularization test=develop

* add comment of function test=develop

* rm name_scope in dygraph mode test=develop

* fix fsp_op, test=develop

* fix fsp grad op maker, test=develop

* update op_use_default_grad_op_maker.spec, test=develop

* fix save_dygraph with nonexistent dir
test=develop

* minor fix
test=develop

* fix unittest
test=develop

* fix static save
test=develop

* fix compile issue, test=develop

* force link libiomp5 when mklml is enabled, test=develop

enhanced ops: concat, nearest_interp, deformable_conv_v1, sequence_conv, transpose2, conv2d

* remove the useless third_party library from C++ inference library

* revert removing the install directory

* refine code, test=develop

* reuse contain_var, test=develop

* add all_parameters for Program, test=develop

* refine unittest, test=develop

* add example print result, test=develop

faster build by reduce by-product, reduce linking library and fix compile warning of std=c++11 (#22164)

This reverts commit c7b03d30.

* add collective communication library in fleet to replace mpi
* test=develop

* add ci approvals for dygraph optest, test=develop

* Add error message, test=develop, test=document_fix

* Increase the number of approves of CI check for dygraph optest, test=develop, test=document_fix

* add ci approvals for dygraph optest, test=develop

* polish code style, test=develop, test=document_fix

* remove ci-check trigger example, test=develop, test=document_fix

- fluid和lite精度比较替换为相对误差

* test=develop, bug fix for sparse recorder

* polish error message of parallel executor, test=develop

* change PADDLE_ENFORCE, test=develop

* add parameter check; test=develop

* change parameter name checker in dygraph guard; test=develop

* fix test layers error; test=develop

* revert some code to develop; test=develop

* fix exampel error; test=develop

* fix comment error; test=develop

* fix comment error; test=develop

- 添加了fluid-lite子图方式运行resnet的单测
- 修改了依赖Lite的git commit id

* add NotImplementedError for multi optimizers used on multi-places . test=develop

* assert error only if num_devices>1. test=develop

* set test_optimizer_in_control_flow in CMakeLists for using multi-GPU.test=develop