bug fix: reduce_all, reduce_any register GRAD_OP, but have not defined GradKernel

* wip simplify operator framework

* wip

* wip

* done test=develop

* clean test=develop

* fix test=develop

* fix deps test=develop

* fix cpu build test=develop

* fix tensorrt build test=develop

* fix tests test=develop

* fix test=develop

* fix cpu build test=develop

* "split into multiple .ccl"

* "refine file structure"

* "refine files"

* "remove the cmakelist"

* "fix typo"

* "fix typo"

* fix ci

* Enhance reduce op for multi dims.

* Uncomment some unitest.

* Uncomment unitest.

* Remove unused code.

* Fix infershape and python wrapper.

* Add more examples.

* Fix l2_normalize.

* Fix normalization_wrapper.

* Polish code.
1. Rename unitest function.
2. Rename const variable.

* Fix CPPLint errors in multiclass_nms, nccl, nce, reduce and save_load_combine

* Fix

There are mainly following fixes:

- take `DeviceContext` as the template parameter of math functors and OpKernel instead of `Place`
- remove `eigen_device` interface in base class  `DeviceContext`
- remove `GetEigenDevice` interface in `ExecutionContext` and base class `DeviceContext`
- remove unused `platform::EigenDeviceConverter`
- rename `REGISTER_OP_GPU_KERNEL` to `REGISTER_OP_CUDA_KERNEL`
- rename `USE_GPU_ONLY_OP` to `USE_CUDA_ONLY_OP`