* Add pass compatibility check for four mkldnn pass, test=develop

* fix bug sequececonv_eltadd_relu_fuse_pass, output error when sequence_conv's padding_start > 0

* fix seqconv_eltadd_relu_fuse_pass unitest error

* pass compatibility

* add compatibility registry

* add unittests for different padding

* add assert

* drop errmsg

* 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>

* optimize errror report

* add test case for pad op converter

* fix some spelling mistake commented by peiyang

* refine infer ir pass tests, test=develop

* commit to change copy_from_cpu input to np array, test=develop

* fix trt fc fuse test, test=develop

* fix trt_transpose_flatten_concat shape, test=develop

* Add three passes and api reference of paddle_pass_builder.h

* add transpose_flatten_concat_fuse_pass tests for gpu and trt, test=develop

* update test_inference_api.py, test=develop

* add InterencePassTest for testing precision of inference passes, test=develop