* 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.

1. If all bboxes are not difficult ground truth, the users can not define the data layer for this flag and not the input can be None for detection_map API.
2. Set default value for aspect_ratios in prior_box API.
3. Add more check in prior_box API.

* Added kernel to is_empty_op

* Added python API

* Updated code as required

* Updated unittests

* listen_and_serv use localscope

* fix ut

* "fix ci"

* "fix network config"

* "rerun ci"

Change default split minimize value to speed up transport data to pserver

* add dist transpiler unit test

* update by comment

* update by comment

* fix ci

* fix ci

* "add float64 tests"

* "fix based comment"

* "fixed based comment"

* Added auto transform to beam_search_decode_op

* Added some comment

* Added unittest for beam_search_decode_op on GPU