Add LRN efficient GPU implement

* Add DataFeeder

A v2 API like data feeder for book demos.
We can feed data directly from reader.

* Fix CI

* Add an unittest for while/rnn op forward

* Add unittest for raw while op backward

* Fix CI

* Complete Dynamic RNN

* Using latex fraction syntax in sigmoid and logsigmoid op

* Fixing the rendering of the latex equations in adamax operator

* add crf_decoding layer

* fix some typo

* init trunk_evaluator

* add trunk_evaluator layer

* update chunk_eval_op and test, change int32 to int64

* fix a numeric problem

* change layers.trunk_evaluator to layers.trunk_eval

* fix typo

* add precision_val

* add crf_decoding layer

* fix some typo

* fix test_crf_decoding_op

* Add DataFeeder

A v2 API like data feeder for book demos.
We can feed data directly from reader.

* Fix CI

* Add an unittest for while/rnn op forward

* Add unittest for raw while op backward

* Fix CI

* Add hinge loss op

* Update hinge-loss equation for proper latex

* fix grpc compile warn

* update

* -Wnon-virtual-dtor -> -Wno-non-virtual-dtor

* change GPU memory allocating policy

* fix potential overflow bug

Add version and commit information in capi config.h and use unofficial glog for Android API < 21. (#6113)

* Automatically configure the version and commit information in capi.

* Use the unofficial glog repository for building for Android (API < 21).

Also, change the `labels` to `label` for api consistency

* Enable the case N != ldc in EigenBlasGemm.

* Use MemoryHandle instead of direct calling of posix_memalign to alloc temporary memory.

* Use Eigen's slice() instead of a temporary memory.

* Add if-else for different cases in EigenBlasGemm (for N ?= ldc).