test=develop

test=develop

* Add InitializeGPUAndLoadModel to solve random hang when downloading sparse parameters.
* Update SaveBase to solve test problem.

* Merge hapi into Paddle

Hapi is a high level API for training and inference.
The main modules include Model, Loss, Metrics, Dataset.
Also includes common modules and models in NLP and computer vision, such as BERT, ResNet.

These modules are developed by:
0YuanZhang0, guoshengCS heavengate, LielinJiang, qingqing01, xyzhou-puck huangjun12, wangxiao1021, zhangyang.

* add new macro BOOST_GET_SAFELY & unittests, test=develop

* add different macro type, test=develop

* fix get macro type in executor, test=develop

* four macro part change backup

* using one macro for all case, test=develop

* revert attribute change, test=develop

* change to three func to solve gcc4.8 bug, test=develop

* polish some details, test=develop

CreateStepScopes in recurrent op also clears scopes, which can cause segmentation fault un multi-threading. We add a lock in this PR but it may slow the computation process. We will fix in another way in next PR.

* raise index error when slice out of range; test=develop

* add uni test; test=develop

* fix format error; test=develop

* add comment for py::index_error; test=develop

* polish error message; test=develop

* polish error message; test=develop

* Enabled quantize all and skip missing in QAT

* support to train in static

* support to independent decorator

* remove in_dygraph_mode condition in ProgramTranslator

* fix import param_guard and add train/eval test=develop

* Modify into ShareVarsFromScope and rm __all__ in partial_program test=develop

1. To make ProgramTranslator to support `assert` grammar, this PR adds `assert` python API and C++ code. 

2. Fix a bug: graph_pattern_detector.h #include <gtest/gtest_prod.h> but didn't declared dependency at CMakeLists, which can cause single build failure.

3. Refactoring `Formatter` in print_op to make it reusable and reuse the formatter to print in assert op.

test=develop
test=win_gpu

* add UT for mkldnn_cache_capacity

test=develop

* fix comparison of integer expressions of different signedness

test=develop

* API/OP (margin_rank_loss, nce, row_conv, glu, positive_negative_pair) error message enhancement, test=develop

* add unittest for glu, test=develop

* Replace dygraph_to_static_func with @declarative or program_translator.get_func in test_list.py

* Add comments in ConditionalBlock.

* Support list pop last item. 

* Support pop the i-th item. 

* Support an empty tensor array as Input in assign op and set the kernel type is float.

* solve mklml memory leak on windows

test=develop

* remove unused msvcr120.dll

test=develop

* improve activation ops performance, test=develop

* use 32bit only GPU computation, test=develop

* add timeout and http store in communication, add revert and confirm in fleet
* test=develop

* Fix cusolver loader for Windows in dynamic_loader.cc. test=develop

* Fix missing CUSOLVER_ROUTINE_EACH_R1.
test=gpu
test=develop

* Add unsupprot for cusolver on Windows temporarily. test=develop

* Fix GetCusolverDsoHandle error message. test=develop

Add the pslib graph batch reader mode, add the test case for this change

* Support LoDTensorArray for slice op.
* Support read elements of list in dygraph_to_static
* Fix infershape add test for infershape.
* Support Tensor for Attr(starts) and Attr(ends). 
* Use new interfaces in VarTypeInference. 

fix kron_op: when only one input needs gradient

* enhance add_position_encoding error message, test=develop

* enhance warpctc & scaled_dot_product_attention error message, test=develop

* modified error message and ctest of scaled_dot_product_attention, test=develop

* update the analysis predictor, test=develop

* update the unit test, test=develop

* no priority set before the inferface determined, test=develop

* interface name generalization, test=develop

* update local data preprocess doc

* update for 2.0 QAT

test=develop

test=document_fix

* update benchmark data

test=develop

test=document_fix
Co-authored-by: NWojciech Uss <wojciech.uss@intel.com>

* remove double registery for pyramid_hash op

* add noavx_axpy and noavx_axpy_noadd

test=develop

test=develop

* add print transformer & unify print format, test=develop

* remove using of dygraph_to_static_func, test=develop

* remove python stdout capture, test=develop

* fix compatibility problems for PY2, test=develop

* fix detail error, test=develop

* fix type analysis bug, test=develop

* fix print tuple compatible error in PY2, test=develop

* replace get_func to declarative, test=develop

* fix detail bug, test=develop

* fix some detail problems, test=develop

* change visit_call in print transformer, test=develop

* add trace op, test=develop

* Optimized the implementation of trace op, test=develop

* fix a bug of include in trace_op.h, test=develop

* move trace API from creation to math, test=develop

* modified en doc. test=develop

* add complex trace api

* add complex sum api, test=develop

* modified en doc of complex sum and trace, test=develop

* modified doc and trace API, test=develop

* modified en doc of trace and sum, test=develop

* modified comment in complex kron API, test=develop

* OP Should Not Have Unused Input, test=develop

* add GetExpectedKernelType, test=develop

c++ API ( average_accumulates, tensor_array_to_tensor and average_accumulates) error message enhance. test=develop (#23631)