* optimize slice TRT plugin

This patch removes unnecessary barrier for data transfer of needed offset,
so data transfer can be overlap with GPU kernel execution.

This patch also fixes incorrect name of slice plugin. That is, replaces
"layernorm" with "slice"

test=develop

* add serialize/deserialize to slice plugin

* add static shape slice trt plugin

* fix slice trt op convertor dynamic shape bug

* fix format by clang-format

* fix pylint format error

* fix problems commented by peiyang
Co-authored-by: NRyan Jeng <rjeng@nvidia.com>

* polish framework error msg part 6

* polish lossed item

* fix failed unittest

* polish by reviewer comments

* optimize errror report

* add test case for pad op converter

* fix some spelling mistake commented by peiyang

* replace sequence length attr to input

* fix trt_dynamic_shape_ernie_deserialize_test

* support when opt cache dir does not exist

* improve err report, test=develop

Enhance the error messages for  files in operators/math

Update amp_check_finite_and_scale_op and add an updating_loss_scaling op for static graph amp training. (#26240)

* update amp_check_finite_and_scale_op for static_amp.

* use amp_check_finite_and_scale in static graph amp.

* update grads to zero when grads own infinite values(as for amp_checkout_finite_and_scale op).

* add update_loss_scaling op in cpp.

* add update_loss_scaling_op unit test.

* update the doc of the check_finite_and_unscale op

* Update the process of gradients updating skipping if the gradients have infinite values.

* update the way to zero grads.

* update test_update_loss_scaling_op.py

* add log info when find infinite grads.

* add the unit test for UpdateLossScaling Layer.

* rm auto from localsgd

* Add int8 GRU kernel with UTs

* Lint fixes

* More lint fixes

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 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

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

update error info for selected_rows_functor

add lars to fleet meta optimizer

* set default value to strategy in distributed_optimizer test=develop

optimize the error message for math dir

* fix frobenius_norm error, rm p=0 2-axis support. test=develop

* fix summary rnn

* fix kl_div bug when input shape is [1] and reduction is batchmean