* [WIP] Add mp_all_reduce asynchronize overlap.

* Fix some problems.

* Fix dw compute bug, and use a temporary solution to achieve overlap.

* Use fused_linear_param_grad_add to compute dw.

* Reformat ColumnParallel _overlap_linear. Use environment flags to
control following behaviors:
1. export Flags_mp_aysnc_allreduce=True to turn on mp async all_reduce
2. export Flags_skip_mp_c_identity=True to skip two c_identity operators
   in dygraph mode.
3. export Flags_fused_linear_param_grad_add to enable fused_linear_param_grad_add
   in ColumnParallel backward with mp async all_reduce.

* Polish code.

* Remove useless communication API.

* Fix some problems in mp_async_all_reduce and skip_c_identity.

* Add test cases.

* Remove environment variable Flags_fused_linear_param_grad_add in test case.

* Reset error threshold.

* Reset threshold in test case.

* Add useful log. Remove useless test cases.

* remove the optimizer base and learning rate base

* fix bug

* fix bug

repacle fluid.io.load_inference_model, fluid.io.save_inference_model in fluid with 2.0 version  (#55345)

* repacle fluid.io.load_inference_model

* replace fluid.io.save_inference_model

* fix some bug

* fix some bugs of load & save model

* fix some bug

* fix test_inference_model_io bug

* fix word2vec_inference_model bug

* fix some bug

* fix valueError bug

* fix some bug

* fix a warning error

* for debug

* for debug

* fix io error

* fix test_wordvec_book error

* remove debug print

* fix load_var bug

* for debug cinn test

* revert cinn & fix inference_pass_test in windows

* fix some bugs

* revert cinn & fix inference_pass_test in windows

* for debug vars

* for debug

* fix quant_dequant_test

* fix some path errors

* remove fluid save/load

* fix incubate-fleet save

* move some from fluid.io to static.io

* move fluid apis

* fix type error

* remove static exponential_decay

* fix some import error

* remove nn.py

* fix some error

* fix type error

remove the AdamOptimizer、SGDOptimizer、MomentumOptimizer、ModelAverage、LookaheadOptimizer、FtrlOptimizer、DecayedAdagradOptimizer、DpsgdOptimizer in fluid and relocate the ExponentialMovingAverage、PipelineOptimizer、GradientMergeOptimizer and change optimizer base for LarsMomentumOptimizer and RecomputeOptimizer (#55970)

* change the optimizer base for SGDOptimizer

* change the optimizer base for SGDOptimizer

* replace the SGDOptimizer with SGD

* fix bug of sgd

* change the optimizer base for MomentumOptimizer

* fix the remaining tests

* remove the Momentum in fluid/optimizer.py

* fix bug

* fix bug

* fix bug

* fix bug

* Update test_resnet_cinn.py

* Update test_resnet_prim_cinn.py

* fix bug

* fix bug

* fix bug

* remove the ModelAverage in fluid

* remove the LookaheadOptimizer in fluid

* fix bug

* remove AdamOptimizer in fluid

* Update test_image_classification_fp16.py

* fix bug

* relocate the ExponentialMovingAverage in fluid

* restore the static api

* remove the FtrlOptimizer in fluid

* remove the DecayedAdagradOptimizer in fluid

* remove the DpsgdOptimizer in fluid

* fix bug

* fix codestyle

* fix bug

* fix bug

* relocate the PipelineOptimizer

* relocate the GradientMergeOptimizer

* fix bug

* fix bug

* fix bug

* fix doc

* Update __init__.py

* Update test_fleet_qat_meta_optimizer.py

* change optimizer base for LarsMomentumOptimizer

* fix bug

* fix conflict

* fix code-style

* fix sample codes

* fix bug

* fix bug

* fix cinn bug

* fix bug

* fix bug

* Update qat_optimizer.py

* Update __init__.py

* fix bug

* change optimizer base for RecomputeOptimizer

* fix bug

* fix bug

* Update test_imperative_optimizer_v2.py

* Call multiply_ instead of scale_ to avoid multiple DtoH copy.

* Call _squared_l2_norm to calculate grad_clip.

* Fix import error.

Signed-off-by: Njjyaoao <jjyaoao@126.com>

* fix new launch

* fix ps uit

* Support AMP program for onnx QAT API

* Integrate QAT into distributed optimizer

* Reduce the size of test data and increase time limit

* Use logger and reduce time limit of unittests

* Rename and move unittest into fleet test

* Test qat_init API

* fix protobuf problem

* fix protobuf problem

* support sharding parallel

* fix name

* fix

* update

* test amp for sharding

---------

Co-authored-by: pangengzheng <pangengzheng.baidu.com>

replace the AdagradOptimizer 、adamaxOptimizer、AdadeltaOptimizer、RMSPropOptimizer、LambOptimizer and Momentum (#54152)

* replace the AdadeltaOptimizer with Adadelta

* replace the RMSPropOptimizer with RMSProp

* replace the LambOptimizer with lamb

* replace the momentum in contrib/optimizer.py with Momentum in python/paddle/optimizer/momentum.py

* fix bug

* fix bug

* fix bug

* fix bug of Lamp

* fix bug of Lamp

* fix bug of import

* replace the AdamaxOptimizer with Admax and change the optimizer base for AdagradOptimizer

* fix bug

* fix bug

* Update optimizer.py

* fix bug

* fix bug

* fix vpp mp init prob

* triger ci

* no endpoints in dy mode

* fix fleet api inconsistent

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* use global group to pass meta

* use batch isend irecv

* add partial send/recv

* remove communication group

* remove p2p on npu and xpu

* remove virtual pp ut