* add to_readable_code method, test=develop

* polish doc details, test=develop

* polish doc note, test=develop

* fix unittest error, test=develop

* fix coverage, test=develop

* add print test, test=develop

* add print param, test=develop

* hidden to_readable_code api, test=develop

* remove original tool methods, test=develop

* remove old api using code, test=develop

* Add two types of Metric Calculator: MultiTaskCalculator & CmatchRankCalculator.
* Add a config for DynamicAdjustChannelNum function to denote whether we will discard the remaining instances when they are not be distributed evenly.
* Remove CPU code in Pull/PushSparse and we will add it back when testing it fully.
* Fix some known issues: such as copying persistable vars after one epoch running.

* add sync communicator and implement

* add deprecated for distribute transpiler, will delete it after 2.0.0, test=develop

* add half_async in the communicator
* fix DistributedStrategy

* test=develop, bug fix for sparse recorder

* add special way to add distribute vars， Update Pyramid hash op

* fix grad clip， clip op belongs to Backward op when running in Parameter Server mode.

* add distributed_strategy

* add fake init for the trainer, fix large memory hold in the trainer
* do not merge recv vars from a remote endpoint, test=develop
* add recv and save op, merge slice var in one op, save memory
* remove hsigmoid with pull sparse, test=develop

* Add a single_process_multi_thread transpiler.
* Add some UTs.
* Fix some API description.

* add Adam beta1/beta2 support Variable. test=develop

* update pserver decay blocks

* update distributed notify handler

* fix fetch handler error with pslib
* fix distributed lookup table op with 1 pserver

* test=develop,Fix communicator slow bug

* test=develop, delete if() in stop_worker()

* test=develop

* fix UT, test=develop

* fix bug in fetch handler, test=develop

* fix bug in fetch handler, test=develop

* test=develop, fix fetch barrier bug

* test=develop, bug fix

* test=develop, bug fix

* test=develop, fix bug

* bug fix: invalid learning rate decay in pserver async mode

* doc fix, test=develop, test=document_fix

* test=develop,test=document_fix,fix transpiler doc,add API.spec

* delete debug vlog & add rpc function & fix word2vec bug & speed GEO-SGD

Heartbeat for distributed async training.

* refector geo sgd & communicator

* add deprecated memory optimize doc, test=develop, test=document_fix

* merge develop to solve conflict, test=develop, test=document_fix

* fix DistributeTranspilerConfig document, test=develop

fix wrong place with distributed_lookup_table

* fleet api add input check, test=develop

test=develop

* fix correctness of the communicator

* fix a bug in send thread when sending var context is empty, test=develop

* add lookup_table_prefetch_op and prefetch optimize, test=develop

* remove remote prefetch GPU supported

* word2vec force with CPU, test=develop

* test dist remote lookup table force with CPU, test=develop

* fix sync mode hang in transpiler
* remove sync mode in send/recv
* replace PADDLE_ENFORCE with PADDLE_ENFORCE_NE

* remove unused inference_transpiler unit-tests

test=develop

* remove InferenceTranspiler usage in quantize_transpiler.py

test=develop

Polish fleet API to support cuda collective mode and nccl2 mode

* deprecate python memory optimize, test=develop

* remove memory_optimize in unittests, test=develop

* add unittests to deprecated interfaces, test=develop

* remove legacy memory optimization codes, test=develop

* follow huihuang's comments,test=develop

* follow luotao's comments, test=develop

* supports distributed classification training
* update API.spec
* fix evenly division in python3
* change "index_range" to "index_num" in shard_index operator
test=document_preview
test=develop

do some odd jobs, test=develop

1. Since allreduce op has 4 reduce types, We split these four reduce types into four ops
2. We also refined the collective op code, e.g. we separated the collective op kernel into CPUKernel and CUDAKernel, and remove the device specified DeviceContext parameter in template as we already knew the target DeviceContext
3. We remove the newly added Collective op role to reduce the complexity of program and graph analysis