Cache the chosen kernel of operators

test=develop

* Try to enable the runtime_context_cache pass in train phase.

* Put the append of runtime_context_cache pass ahead of multi_dev passes.
test=develop

reduce all analyzer_test ci elasped time

comment np.array(x.get_tensor()) in imperaitve mode to avoid OOM.

fix dataset bug

test=develop

* add unit test for fuse_opt_ops
test=develop

test=develop

test=develop

test=develop

Imperative test layers without parameters

test=develop

Revert "Model data cryption link all lib"

test_analyzer_int8 tests use default pass order

test=develop

test=develop

test=develop

add NCCLParallelContext for parallel dygraph

test=develop

test=develop

test=develop

test=develop

* add pixel_shuffle op

* add pixel_shuffle op, test=develop

* rewrite code, test=develop

* delete useless comment, test=develop

* Refine pixel_shuffle_op and unit testing

* refine code,test=develop

* refine .cu,test=develop

* fix unittest,test=develop

* Fix unit testing
test=develop

* resolve conflict, test=develop

* fix test, test=develop

* fix API, test=develop

* fix test datatype bug,test=develop

* polish comments,test=develop

* add API,test=develop

* test=develop

* Add Pixel_Shuffle OP,test=develop

* support python3,test=develop

* add include memory to travis CI bug,test=develop

test=develop, fix transformer in dygraph

test=develop

fix load bug about context.place

test=develop
This reverts commit c38c7c56.

* Fix bug of gradient interface

* shrink transformer

* Right transformer

* Change from width-first backward to deep-first backward process

test=develop

* Reverse iterator op's input

test=develop

* Polish code

* Change the iteration direction in ingrads' map slots

test=develop

* Polish code

test=develop

* Add GPU place in static mode of ptb rnn

test=develop

* Polish code

test=develop

/

add some dygraph op
1.Conv3d
2.Conv3dTranspose
3.RowConv
4.GroupNorm
5.SpectralNorm
6.TreeConv

and utest