Merge branch 'int8' of https://github.com/chenjiaoAngel/Paddle-Lite into int8

pull code

 fix reshape infer shape issue  
adaptive pool, 
support adaptive pool2,  
multi thread ok,   
optimize global pool, 
support faceboxes and behavior image, 
realize bm device info, 
multi device ok, 
support multi cards, 
support efficienet

* split subgraph model doc

* remove lite_with_clang, test=develop

* fiix format, test=develop

* [LITE][OPENCL] add gpu perf mode, priority level for qcom adreno. test=develop

[OPs] [ARM Kernel]move assign_value op into basic ops, while moving fake ops about quantization into extra to balance lib size (#3518)

* fix:modify quant_op to extra and add one basic op&kernel

* fix: move fake op to extra

* add interpolate type2 pass. test=develp

*  Add shufflenetv2 in test_libs demo, test=develop

* [LITE][OPENCL] Fix opencl miscs. test=develop

* [LITE][OPENCL] fix opencl api. test=develop

* fix. test=develop

* [LITE][OPENCL] Fix OpenCL activation judge of Conv Kernel. test=develop

* change back for cxx_api.cc test=develop

* change back for cxx_api.cc test=develop

* Fix activation act_type default value. test=develop

* Fix enum value. test=develop

* Fix conv test for relu. test=develop

* [LITE][OPENCL] Fix input randaom range of conv1x1 opencl kernel. test=develop

* fix miscs. test=develop

* add test for imfix ,test=develop

* fix group like 3x3,test=develop

* add direct release build script, test=develop

* fix group like conv , test=develop

* add test for imfix ,test=develop

* fix group like 3x3,test=develop

* add direct release build script, test=develop

* add test for imfix ,test=develop

* fix group like 3x3,test=develop