* Incorporate cudnn_lstm into LSTM api.
test=develop

* Make coalesce_tensor support alignment optionally.
test=develop

* Reorganize RNN apis. test=develop

* Fix cudnn rnn layout conversion.
test=develop

* Add sequence_length support for RNN cudnn implement.
Add optional init_h and init_c gradient for cudnn_lstm_op.
test=develop

* Use create_parameter for rnn cudnn impl.
test=develop

* Move `self._flat_weight = self.create_parameter()` in RNNBase to main_program.
test=develop

* Update RNN api unittest to use set_device.
test=develop

* Fix set_place for unit tests of RNN apis.
test=develop

* Fix use_align in coalesce_tensor_op.
test=develop

* Adjust RNN apis arguments according to comments.
test=develop

* Polish documents for SimpleRNN apis.
test=develop

* Refine random seed in cudnn_lstm_op.
Expose rnn params from sublayers to RNN.
test=develop

* Fix RNN saving for jit.save.
Refine cudnn_lstm dropout behavior.
test=develop

* Fix doc of GRU. test=develop

* Use ShareDataWith to avoid copying for cudnn_lstm_op test.
test=develop

* Remove updates on cudnn_lstm temporarily.
test=develop

* Use ShareDataWith to avoid copying for cudnn_lstm_op test.
test=develop

* Refine random seed in cudnn_lstm_op.
test=develop

* Fix test_lstm by adjust ConcreteProgram buffer getter.
test=develop

* Use create_parameter instead of create_var for rnn._flat_weight for static graph usage.
test=develop

* Remove W input for cudnn_lstm to pass unused_var_check.
test=develop

* Add test_predict for RNN unit tests coverage.
test=develop

* Fix code style of rnn.
test=develop

* Fix F.rnn usage in rnn.py.
test=develop

* test=kunlun;

Add elementwise XPU OP kernel for KUNLUN core, including (but still cannot process common broadcast):

    * elementwise_div op
    * elementwise_max op
    * elementwise_mul op (with grad op)
    * elementwise_sub op (with grad op)

* 0.05->0.01

* add xpu error message description;test=kunlun

* conv dilated mkldnn support: forward and backward pass

* add mkldnn conv_transpose dilation UT
test=develop

* remove unnecessary PADDLE_ENFORCE

* add int8 and bf16 dilated conv UT

* update according to reviews

* add size method for large scale

* add large scale UT

* add ut for checkpoint

* test=develop, fix geo sgd communicator

* test=develop, gloo_init_method

* test=develop, bug fix for gloo http_init

* add reduce xpu op test=develop;test=kunlun

* add reduce xpu op test=develop;test=kunlun

* add reduce xpu op test=develop;test=kunlun

* add reduce xpu op test=develop;test=kunlun

* add reduce xpu op test=develop;test=kunlun

* add tensor clone

* fix unittest test_var_base

1. support channel last in BatchNorm*d (#27875)
2. fix a bug in batch_norm_op cuda kernel by extracting ResizeToChannelFist(Last), TransToChannelFirst(Last) to operators/layer_utils.h

* support setting xpu place

* add ut, test=kunlun

* fix adam

* fix gpu adam

* fix code style

* fix ut

* update ut add cuda code

* add stack pool2d roi_align xpu op,test=kunlun

* error message opt, test=kunlun

* add xpu unittest,test=kunlun

* skip check grad,test=kunlun

* fix boostget , test=kunlun

Add elementwise XPU OP kernel for KUNLUN core, including (but still cannot process common broadcast

* add sparse tensor load method

update the code for the topk message optimize 

* polish some error message

* add white list

* revert shell script change

* test=develop, bug fix for parameter_recv
* test=develop, for unittest, test_fleet_rolemaker_new

*test=kunlun

* add multitask

* add multitask, test=develop

* fix code style, test=develop

* add partail push dense, test=develop

* fix has_kay in py3, test=develop

* fix, test=develop

* fix, test=develop

* fix, test=develop

* fix norm api doc, test=develop

* fix error message, test=develop

* fix api norm, test=develop

* add adagrad, test=develop

* fix bug, test=develop

* fix bug, test=develop

* add spetral_norm, test=develop

* fix adagrad, test=develop

* merge , test=develop

* add lookup_table_v2_op_xpu, test=kunlun

* add lookup_table_v2_op_xpu, test=kunlun

* change some Tips ,test=kunlun

* use exhaustive_search for float16

* tune algo only when dtype is float16

* add dropout,log_loss, test=kunlun
* fix dropout, test=kunlun
* polish error message, test=kunlun
* change boost::get to BOOST_GET_CONST, test=kunlun
* fix copyright, test=kunlun

* support mean,softmax_with_cross_entropy on Baidu Kunlun,test=kunlun

* fix unittests error,test=kunlun

* delete boost::get,test=kunlun

* add xpu sgd & momentum

* add conv for xpu, test=kunlun

* polish error_message, test=kunlun

* polish error_message, test=kunlun

* fix copyrigth, test=kunlun

* add xpu slice op
test=xpu

* add slice xpu op
test=xpu

* code style
test=kunlun

* style
test=kunlun

* format
test=kunlun

* op error info

* style

* code format

* add gather_op xpu, test=develop, test=kunlun

* fix ut, test=develop, test=kunlun

* fix the ut,test=develop, test=kunlun

test=develop

* add load_op_xpu for Baidu Kunlun, test=kunlun

* add is_compiled_with_xpu for unit test, test=kunlun

* add is_compiled_with_xpu for unit test, test=kunlun

* use IndexList to improve performance, test=develop

* remove EIGEN_HAS_INDEX_LIST, test=develop

* use IndexList only when EIGEN_HAS_INDEX_LIST is true