[LITE][COMPILE] Fix clang compile slow for arm cpu and abnormal exit for opencl. test=develop (#3763)

* [LITE][COMPILE] Fix clang compile slow for arm cpu and abnormal exit for opencl. test=develop

* add note for compilation option. test=develop

* fix pooling overflow, test=develop

* fix conflict test=develop

* [LITE][OPENCL] Fix grid sampler for mali gpu. test=develop

* [LITE][OPENCL] Fix OpenCL compatibility for opencl driver 1.10. test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop expend opencl kernel & unit test ,test=develop

* [OPENCL] develop expend opencl kernel & unit test ,test=develop

* [LITE][PROFILER] Fix unit test segfault when profiler on. test=develop

[LITE][OPENCL] Add local work size, global work size for opencl profiler; fix bug for variable inference pass. test=develop (#3741)

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

int8 direct_conv, dw_conv add relu6 and leaky relu fusion

add a CI rule:  You must have Superjomn approval for change 20+ files or add than 1000+ lines of content  (#3727)

* test=develop, add a CI rule

* test=develop, remove the debug messages

* improve 3x3s1 direct profile

* add gemv+relu6/lleakyRelu

* fix relu6 problem, test=develop

* fix format, test=develop

* add six / scale , test=develop

* [LITE][PROFILE] Fix summary info for profiler. test=develop

* reduce .so size. test=develop

* compile all targets when LITE_ON_TINY_PUBLISH=OFF

* unordered_map is more convenient when key is customized class

* test=develop

* reconstruct op_registry. test=develop

* remove function GetKernelOffset. test=develop

* [LITE][DOC] Add kernel profiler doc for add operation. test=develop

* add argmax profiler info. test=develop

* fix pooling bug and speed

* add 2x2s2p1 pooling. test=develop

* fix conflict, test=develop

* add calltime/gops for summary. test=develop

* fix act case for kunk. test=develop