* [Custom Op]Remove PADDLE_WITH_MKLDNN in custom_op  (#32903)

* [Dy2Stat]BugFix StaticAanlysis with gast.Subscript (#32969)

* BugFix StaticAanlysis with gast.Subscript

* remove codes

    test_paddle_save_load 单测随机挂：使用np.ndarray生成随机数组，可能生成nan，造成做对比时结果不匹配（nan != nan）。改为np.random.randn生成随机数组。

    test_paddle_save_load_binary随机挂: 如果一个字符串不能解析为Program，windows上会有超时风险。解决方法：不在windows平台不加载'不能解析为Program的字符串'。

原始PR:#32949

Change Paddle CI-Cverage Python3.8 Cherry-pick 32515

cherry-pick of #32972

* cherrypick for #32640 :add profile and fix dataset hang in heterps;test=develop

* cherrypick for #32640 :add profile and fix dataset hang in heterps;test=develop

* cherrypick for #32640 :add profile and fix dataset hang in heterps;test=develop

* doc of paddle.save/load

* polish doc of paddle.save/load

 Fix CI Python3 on release/2.1 #32930 

* fix dataloader exit hang when join re-enter. test=develop

* double check _shutdown. test=develop

fix error log for reducer

fix doc

fix bug of utest

fix spawn

fix converage

* implement MHA order same as training

* fix fp16 compile issue on old architecture
Co-authored-by: Nzlsh80826 <rewang@nvidia.com>

* remove packages in __all__

* create new public api level paddle.callbacks;paddle.hub;paddle.utils.unique_name

* fix compile error on jetson platform

* remove unused head file

* rm decode_jpeg op on jetson platform

 [CHERRY-PICK2.1]Remove paddle_custom_op dynamic libraries, and link to FLUID_CORE on windows (#32583) (#32769)

* Remove paddle_custom_op dynamic libraries, change link to FLUID_CORE on windows, and check copy_to

* fix CI

* Add casting initializers for bf16 training

* Changes after review

* Correct test and add comment
Co-authored-by: Njoanna.wozna.intel <joanna.wozna@intel.com>

* Add deprecated warning info.

* Add unittest for deprecated decorator.

* Add warning info for tensor.grad

修复了py_layer_op由于没有析构PyLayerContext造成内存(显存)泄露的问题。

原始pr：#32707

* clear 'BasicEngine' when an exception occurs in the backward. (#32546)

* clear 'BasicEngine' when an exception occurs in the backward.

* deal with conflict.

* deal with conflict.

* forward return any type. (#32661)

* revert data_generator

* add setup.py

* base changes for fix

* minor change

* fix for bwd kernel

* removed unnecessary import

* implemented reviewers suggestions

* CI fix

cherry-pick:change softmax_with_cross_entropy_op's parameter name from softmax_switch to use_softmax (#32750)

* change parameter name from softmax_switch to use_softmax, test=develop

* cherry-pick:change parameter name from softmax_switch to use_softmax, test=develop

avoid polluting logging's root logger

* add to_grayscale, normalize

* add rotate

* add vfip and hflip

* add crop center_crop

* add utils

* add utils

* update utils, add raise for some cases

* add padding, support constant, reflect, replicate, circular same as paddle.pad

* update rotate

* using utils func in [v|h]flip

* add get-image-[n,c,w,h] axis utils

* add get-image-[n,c,w,h] axis utils

* align

* update

* remove default value in utils func

* add assert for pad

* update assert paddle image

* support rotate fill func

* raise valueerror for pad

* remove typing, py2 dont support

* init uinttest for transforms tensor

* add resize op

* register [normalize hflip crop center_crop resize transpose] imagenet

* register [normalize hflip crop center_crop resize transpose] imagenet

* fix bugs, (w, h) getter and import

* add _get_image_size for tensor image

* add pad vflip for tensor image

* add unittest for tensor transforms

* update transforms unittest for converage CI probelms, test=develop

* update

* update

* update

* fix `get_shape` for tensor backend

* update

* update

* add more resize tests

* update

* update for ci test

* update

* remove redundancy code

* update uinttest, and set tensor image to hwc by default

* add tensor backend

* fix copyright doc, rm comment code, add pil unittest

* update data_format to `chw` for tensor

* coverage notest,test=coverage

* update

* update

cherry-pick #32561

Dy2stat failed when user writes return paddle.to_tensor(xxx), the reason is that visit_Expr doesn't work when the Expr is in return. Some other statements may trigger same bug. To fix it, we re-wrote a transformer to transform paddle.to_tensor to paddle.assign for all Call nodes.

* update, test=develop