double grad support for abs 

* add batch norm double grad, test=release/1.8

* update, test=develop

* fix, test=develop

* optimize slice TRT plugin

This patch removes unnecessary barrier for data transfer of needed offset,
so data transfer can be overlap with GPU kernel execution.

This patch also fixes incorrect name of slice plugin. That is, replaces
"layernorm" with "slice"

test=develop

* add serialize/deserialize to slice plugin

* add static shape slice trt plugin

* fix slice trt op convertor dynamic shape bug

* fix format by clang-format

* fix pylint format error

* fix problems commented by peiyang
Co-authored-by: NRyan Jeng <rjeng@nvidia.com>
Co-authored-by: NShang Zhizhou <shangzhizhou@baidu.com>
Co-authored-by: NRyan Jeng <rjeng@nvidia.com>

test=release/1.8

* Fix test_gast_with_compatibility.py due to the problem of gast in python3.8 (#27433)

test=develop

* fix dll load bug on windows from python3.8 (#27324)

* Support python3.8 (#26850)

* Support python3.8

test=notest

* Replace the 'spawn' start method with 'fork' start method for multiprocessing, on MacOS with python>=3.8 (#27317)

* Replace the 'spawn' start method with 'fork' start method for multiprocessing, on MacOs when python>=3.8

test=develop

* Correct the error in decorator.py (#27409)

test=develop
Co-authored-by: NZhou Wei <52485244+zhouwei25@users.noreply.github.com>

* add DISTRIBUTED_TRAINER_ENDPOINTS env for cloud

* add double grad for expand, test=develop

* add double grad to reduce sum

* delate some wrong message test=develop, test=document_fix

* delate some wrong message

* add mock barrier all (#24786)

* add mock barrier all
test=develop

* fix
test=develop

* fix
test=develop

* fix
test=develop

* fix gloo error
test=develop
Co-authored-by: Nxujiaqi01 <173596896@qq.com>

* fix dataset py3 (#25012)

* fix dataset py3 error
* test=develop

* fix logger (#24682)

* fix logger of FetchHandler，which may print log twice
* test=develop

* add timeout and http store in communication (#23436)

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

* modify datanorm op test=develop (#23030)
Co-authored-by: Nxujiaqi01 <173596896@qq.com>
Co-authored-by: Nyaoxuefeng <yaoxuefeng@baidu.com>

* fix prelu for compatibility with saved model of old version

* reshape alpha

test=develop
Co-authored-by: Nchalsliu <45041955+chalsliu@users.noreply.github.com>

* fix cpuid.h not found

* fix-jetson-compile-pyramid_hash

* add more info to version.txt, test=develop (#24551)

* Fix dygraph grad bugs (#25781)

* fix double grad visitid unit; test=develop

* change name hash_pair to HashPair; test=develop

* follow comment; test=develop

* remove manual seed; test=develop

* change create_graph from True to False; test=develop

* Fix the double grad bug for the star gan. (#25655)

* update the retain_graph parameter doc. test=develop

* Saving inference model for user defined quantization

* Saving inference model for user defined quantization

* add scale_x_y for yolo_box, test=develop

* add scale_x_y for yolo_box, test=develop

* refine eps in iou_similarity, test=develop

add try..catch...

* add user defined func test=develop

* update test=develop

* update test=develop

* fix name conflicts test=develop

* add unittest test=develop

* change 2018 to 2020 test=develop

* add comment test=develop

* add comment for function test=develop

* fix details test=develop

* fix details test=develop
Co-authored-by: NLiufang Sang <slf12thuss@163.com>

* attempt to resolve tls problem, test=develop

* add glibc version check, test=develop

* fix regex, test=develop

* refine get_libc_ver, test=develop

* refine get_libc_ver, test=develop

cherry-pick,fix optimizer.state_dict and LRScheduler.state_dict to save/load dygraph

add new API:LambdaDecay,test=develop 

* enable load_program_state run in imperative mode; test=develop

* remove useless code; test=develop

* make default_collate_fn visible. test=develop. test=release/1.8

* test=release/1.8, cherry fix fluid.embedding

test=release/1.8

* CHERR-PICK1.8,add base class of LearningRateEpochDecay, and API: MultiStepDecay, and API: StepDecay,test=release/1.8

* fix unittest to add coverage,test=develop