disable seqpool concat pass by default saving CI time

* bug-fix, test=develop

* tweak code, test=develop

list of fixed ops:
lookup_table_op
space_to_depth_op
squared_l2_distance_op
squared_l2_norm_op
teacher_student_sigmoid_loss_op
tree_conv_op
warpctc_op

test=develop

reduce hasAttr elapsed time in RunImpl

test=develop

* make less batch of tests to fit ci

* test=develop, invoke ci

add some comments back

* test fuse all_reduce

test=develop

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