* fix code bug
test=develop

test=develop

* add event for fast executor and add threads for scopebuffer executor
test=develop

Async exe support communicator

test=develop

* add record_event
test=develop

* remove csp
test=develop

Refine the Executor when the num_thread=1

* make parallel_executor support FLAGS_use_mkldnn

test=develop

* add warning when set mkldnn_enabled_op_types_ in non-mkldnn env

test=develop

* reset drop local scope counter
test=develop

test=develop

* move pass to ir

* polish code
test=develop

* fix dependency
test=develop

* add use_cuda to inplace pass,test=develop

* add test softmax_with_xe_inplace test,test=develop

* fix potential inplace bug
test=develop

* add more skip vars in mem opt pass,test=develop

* follow comment,test=develop

* follow comments,move duplicate out arg check to program->graph,test=develop

* add use_cuda to inplace pass,test=develop

* add test softmax_with_xe_inplace test,test=develop

test=develop

* fix op graph view
test=develop

* rewrite inplace pass and fix reference count pass bug
test=develop

* fix unittest failed
test=develop

* follow comments, test=develop

test=develop

test=develop

test=develop

* Add fuse momenutum ops

* speedup gc and inplace softmax_with_cross_entropy_grad
test=develop

* refine models gpu mem
Merge skip vars and warning messages of mem opt
remove relu mem opt
test=develop

* follow comments
test=develop

* fix bug of fuse optimize ops

* replace ThreadedExecutor with FastThreadedExecutor
test=develop

* Fix Travise CI
test=develop

* Test FastThreadedSSAGraphExecutor
test=develop

* refine parallel_ssa_graph_executor.cc
test=develop

* Add an option to enable the cache of expected kernel in train phase.
test=develop

* Change the default value of cache_expected_kernel to true.

test=develop

* bug-fix, test=develop

* tweak code, test=develop

* test fuse all_reduce

* 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

* expose fuse broadcast ops

* Fix the bug of the fast threaded executor. I

* fix the bug of reusing different types of variables in memory_optimize_pass, test=develop

* disable SELECTED_ROWS AND LOD_TENSOR_ARRAY reusage, test=develop