* use global conv cache; test=develop

* use singleton cache; test=develop

* fix format error; test=develop

* add cudnn helper header; test=develop

* fix header error; test=develop

* fix mac unitest; test=develop

* fix mac unitest; test=develop

* fix file format; test=develop

* fix include file error, test=develop

* remove kernel_configs_ in class ExecutionContext and kernel_configs_map_ in class OperatorWithKernel, test=develop

* fix test_elementwise_mul_op_dim, test=develop
Co-authored-by: Nphlrain <phliuhongyu@126.com>

This reverts commit dc229b41.

* add condition judgement for performance improvement test=develop

* add condition judgement for performance improvement test=develop

* refine code style test=develop

* fix_conv_grad_grad

* fix_bug, test=develop

* Delete PadFuntion, include padding.h instead. test=develop

* move function(IsSymmetricPadding) from conv_cudnn_op.cu/conv_transpose_cudnn_op.cu to padding.h, test=develop

1.support asymmetric padding;
    2.support padding algorithm:"SAME" and "VALID";
    3.support channel_last: data_format NHWC and NDHWC;
    4.change doc of python API and c++;

    test=develop, test=document_preview