* [NPU] apply npu_identity to conv bn and copy2cpu, test=develop

* update npu identity to share data with x, test=develop

* address review comments, test=develop

* remove flatten in nn.py under fluid

* fix test case

* fix test case

* fix codestyle

* fix codestyle

* fix code style

* remove extra test case

* remove test case

* remove fluid

* update public

* core

* public

* public1

* ci

* refactor: move wait

* refactor: move barrier

* fix: fix incorrect import

* add pattern for auto search

* add unittest

* add matmul_flops

* add dropout,layer_norm ...

* add dropout,layer_norm ...

* add elementwise_flops

* add dict check

* add unitests

* add equation for flops computation

* regularized annotations

* clear fluid api: warpctc, nce, identity_loss

* fix test_layers.py __init__.py

* fix loss.py

* change __init__.py and api calling method

* fix nce

* fix nce

* fix fluid.data

* delete warpctc api document

* fix loss.py

* fix ctc_loss

* fix test_warpctc_op.py

* fix test_layers.py

* fix some bug

* fix conflict

* fix ci bug

* Empty Commit test=allcase

* fix ci bug

* add_gather_nd_

* add_gather_nd_

* add_gather_nd_

* remove fluid.reduce_sum

* remove fluid.reduce_sum

* modify axis and import paddle

* modify keepdim and out_name

* modift unittest

* modift unittest

* modify CI_static and loss.py

* modify test_mse_loss

* modify static ci

* modify static ci datatype

* add import paddle in test

* fix conflict

* fix conflict

* modify ci

* modify ci

* fix_conflict

* fix bug

* code_style

remove random_crop which is not used in Paddle 2.0

* support xpu scalar inplace

* sharding for xpu

* update

* update
Co-authored-by: Nheyanru <81976792+heyanru01@users.noreply.github.com>

* refactor: move all_gather

* add isort config

* isort all files

* group norm fp16 support

* Clean fluid resize_linear API

* Clean fluid image_resize_short API

* add image_resize back

* attr ready

* op ip ready

* start dynamic

* end2end ok

* input shape to map, stat by op

* layer wip

* first version ready

* fix proto depds

* fix profiler deps

* fix flops typo, rm tuple shape

* support xpu scalar inplace

* sharding for xpu
Co-authored-by: Nheyanru <81976792+heyanru01@users.noreply.github.com>

* add index sample fp16 support

* remove fluid APIs in distributed_strategy.py and role_maker.py

* Revert "remove fluid APIs in distributed_strategy.py and role_maker.py"

This reverts commit 223bbee990d3bf69e252fc3c0f19e3873550a264.

* remove fluid APIs in distributed_strategy.py and role_maker.py

* remove index sample op changes

* remove fluid APIs under fleet.base

* remove fluid APIs under fleet.layers.mpu

* remove fluid APIs under fleet.meta_optimizers

* fix fluid error

* fix util_factory.py

* reset fluid.io.load_inference_model API

* move conv3d

* remove unused import

* remove brelu in nn.py under fluid

* add brelu op test case

* support default use_gpudnn=True

* fully support cudnn in phi

* add header file

* add white_list, verify accuracy

* phi support all cudnn

* opt affine_grad

* try different arches of pretrained_model

* try different arches of pretrained_model

* add debug string

* debug eager_method

* add debug string, pass all local ctest

* polish all debug code

* delete use_cudnn relevant code autogen

* fix depthwise_conv2d

* Share all other members of Tensor except use_cudnn

* polish codes according to review opinion

* polish codes according to review opinion, fix bug

* polish codes according to review opinion, opt performance

* polish codes according to review opinion, fix pooling.py

Note: this is a temporary solution, should be replaced once reduce kernel
is natively supported on KL2

* remove expand and eye under fluid.layers

* delete expand API test case

* do not calc reduce_all in eager mode

* refine python c cast list

* refine

* refine

* refine

* refine

* refine

* refine

* refine

* refine

* refine