* Adapt to batch_norm_grad op and add align function in
roi_align op for kunlun, *test=kunlun

* Adapt to batch_norm, batch_norm_grad op api for kunlun, and add unit-tests of batch_norm, roi_align. *test=kunlun

* change Vector to std::vector and provide MixVector class as a helper wrapper class

* solve the multi-gpu hang problem

* remove the duplicate template instantialize

* Copy vector to cpu

* add CopyToCPU

* xxx

* final version: fix the problem of all reduce

* remove mixvector dependence

* fix

* merge

* fix code

* fix by CI

* fix benchmark and communicator config

* fix bugs of the_one_ps

* multi program and fix bug in optimizer

* multi program in the_one_ps

* public commcontext

* [Phi] Migrate unfold_op into phi

* fix im2col CPUContext template instantial

* fix unfold_op.h header include problem

* fix unittest

* fix PT->PD

* add align for WorkQueue

* add spinlock

* merge develop

* merge

* Add EventsWaiter

* Revert "Add EventsWaiter"

This reverts commit e206173aa9be7401b83a53581627bfaf557c8fb2.

* add log for Executor

* update the profiler
Co-authored-by: Nliutiexing <liutiexing@google.com>

* fix pass and convert_op for preln_ernie and add preln_ernie'flag in pass

* [Auto Parallel] Add the high-level Engine API

* Update the test cmakefile

*test=kunlun

* update. test=develop

* update. test=develop

* fix. test=develop

* update. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update. test=develop

* update. test=develop

* Modify the implementation of BlockYReduce to fit more scenes

* fix

* fix

* add check for using HostAlloc

* add check for using HostAlloc

* Support NoNeedBuffer for final state codegen

* Replaced pten with phi

* Modified RandomKernel with Kernel Primitive API

* update pten.h to phi.h

* update

* update fullKernel

* Add Sort API for Kernel Primitive API

* update & -> ptr

* unify register macro

* rename declare macro

* fix infrt error

* [PluggableDevice]custom kernel to pten core structs

* mod extension.h for custom op

* compatible python for CI

* support custom context

* refactor to pten

* fix windows and ut

* Move Abs InferShaper to phi

* Fix CI error

* migrate huber_loss into phi

* migrate infershape

* modify pten into phi

* Add loss conversion from uint16 to float in progressbar class

* Fix test coverage

* Actually fix coverage

* Fix format error

* fix cond grad error when handle tensor array

* add UT

* gpu ps graph engine

* remove logs

* rm reduce_sum raw kernel

* remove reduce_mean kernel

* remove reduce_mean kernel

* reduce support int and int64_t

* mean support int and int64_t type

* disable some distribute test case when in CPU test env, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop

* fix fill_constant bug, *test=kunlun

* fix fill_constant bug,*test=kunlun