Error description optimize for math dir

* Fix the lookup_table_v2 failed on GRU mkldnn kernel issue
test=develop

* fix according to reviews, removed x_num_col_dims
test=develop

* update gru model. change according to reviews
test=develop

* change according to reviews
test=develop

* Add details to the summary and test it, test=document_fix

* Add set +e before example, test=document_fix

* Remove test code, test=document_fix

* Optimize summary information and test it, test=document_fix

* Remove test code, test=document_fix

* temporally disable zero_copy

* add test

* follow comments

* fix calcu_gradients

* fix code place

* fix embedding interface usage

* support to_static(model)

* add warning and unittest

* Use the single GPU card to execute the test_fuse_bn_act_pass UT.

* add double grad for tile, test=develop

* add double grad for expand_v2 op, test=develop

* add double grad for expand, test=develop

[UT] fix run type of ut test cases of test_train_recognize_digits and test_api_impl, test=develop (#27218)

* - introducing oneDNN 1.6

test=develop

* - Removed redundant code

test=develop

* add some unittest cases ot verify jit.save, no_test

* add more unittests

* add test with example inputs

* polish implement details

* remove useless blank

* fix fetch random error

* parameter_server_optimizer support auto_strategy

 fix the CudaPinMemory bug for the equal op and add the test case for the equal op

Decrease the number of running iterations to reduce CI time.

CI system shows it decreased the unittest time from about 90 seconds to about 30 seconds

update error info for selected_rows_functor

add lars to fleet meta optimizer

* refine launch and distributed repr string for print

* set default value to strategy in distributed_optimizer test=develop