* v2.6 supports mask_demo, test=develop

* add crop
* add crop ut. test=develop
* fix format,test=develop

*  Add shufflenetv2 in test_libs demo, test=develop

* fix cuda demo. test=develop

* update type error. test=develop

* add train_demo, test=develop

* Ehance mobilenetv1/v2, add std_dev computation. test=develop

* Ehance cxx mobile_light demo, add std_dev computation. test=develop

* [LITE][OPENCL] Change fp32 fc to fp16's. test=develop

* fix act in conv3x3opt opencl kernel. test=develop

* [LITE][CMAKE] fix cmake for opencl tiny_pubish compile. test=develop

- 增加cuda c++ demo.
- 考虑到检测模型尾部一般是multiclass_nms，该kernel为host，如果fetch kernel为cuda的话，则会在此处插入无用的io_copy(host->cuda)，由于该原因，注释掉fetch的cuda kernel. 默认使用host的fetch kernel. 此处暗中进行的行为：每次predictor run完，都会默认把数据从cuda拷贝到cpu

* improve conv_dw profile with rel relu6 leakyrelu, test=develop

* add depthwise, test=develop

* fix ci error, test=develop

* fix cv demo print, test=develop

* [LITE][OPENCL] Change fp32 fc to fp16's. test=develop

* fix act in conv3x3opt opencl kernel. test=develop

* [LITE][DEMO] fix input arg. test=develop

* fix shutdownlog for std::endl. test=develop

* [LITE][DEMO] Enhance mobile_light demo. test=develop

* fix print. test=develop

* fix clang v7 error
* change ut, test=develop
* fix conv ut, test=develop

* fix

* fix format. test=develop

* fix format, test=develop

* fix resize error

* fix format. test=develop

* fix format, test=develop

* fix format

* note log(4) << x_image. test=develop

* note x_image printf. test=develop

* fix leakyrelu and tanh compute error, test=develop

* delete extra file, test=develop

* reset act

* fix conflict and readme, test=develop

* fix ios run error, test=develop

update mask detection demo

* make full_publish to fix link bug in mask demo

Update mask demo docs, add model download from paddlehub

* [DEMO] add mask detection demo, test=develop

* [DEMO] fix ssd detection demo bug, test=develop

* FIX CXX DEMO MAKEFILE TO SUPPORT MAC COMPILE test=develop

* test=develop

* add cv image process

* fix arm liunx build error

* add LITE_WITH_CV defien to make cv, test=develop

* fix cv format, annd add describe in utils/cv

* set LITE_WITH_CV=OFF in build.sh, test=develop

* delete cv_enum.h in utils/cv, push the contents in cv_ennum.h to paddle_image_preprocess.h, test=develop

* according to reviews to redefine paddle_image_preprocess.h, test=develop

* add detailed note of flipParam, test=develop

* fix format in paddle_image_preprocess.h, test=develop

* fix cmake error in llite/CMakeLists.txt, missing mkdir cxx, test=develop

* according to review change, test=develop

* add elemetnwise mul constant elimination and deconv+relu, deconv+batchnorm fusion, test=develop

* fix format, test=develop

* fix model_optimize bug, update concat and split op, speed up, test=develop

* update split speed, test=develop

* fix format, test=develop

* add classify demo inn demo/cxx/ , test=develop

* fix formart inn mobile_classify, test=develop

* delete some note and extra code, test=develop

* remove test.jpg and labels.txt, test=develop

* add test_cv in cxx/demo

* add test_cv READMEE, test=develoop

* add note info, flip only support x, y, xy;rotate only support 90, 180, 270; test=develop

* fix build error, paddle_cv_arm , test=develop

* add GRAY to RGBA(BGRA) convert and RGBA(BGRA)_to_Tensor, test=develop

* fix format from review, test=develop

* fix makefile format. test=devellop

* fix bbuuild v7 error, test=develop

* [Demo] add cxx mobilenetv1-ssd detection demo, test=develop

* add makefile to mobile detection demo, test=develop

* [Demo] add cxx mobilenetv1-ssd detection demo, test=develop

* [demo] fix mobile_detection code style, test=develop

* [Demo] fix demo code style, test=develop

* [Demo] fix detection demo makefile dependency, test=develop

* Add CheckInput. test=develop

* Fix android_log_ undef when using .a. test=develop

* add guide of how to use shared library in demo. test=develop