* wip

* add ref_by_trainer_id op

* ready to test

* fix ref inputs

* refine rpc_op_handle

* fix merge bug

Do not use ctor

* Reduce line of codes.
* We can use virtual function for Maker now.
* The implementation does not care what maker holds, it is easier to
refactor later.

* init

* add some check

* add dist transpile logic

* add insert op for block

* init change get_pserver_program

* optimize code

* fix a bug

* can run now

* start to do table split

* start to process table gradient

* complete pserver part

* can send_vars now

* revert cpplint

* fix a bug

* optimize code

* move dist test to models

* revert the interface of distribute_transpiler.transpile

* fix prefetch_block

* optimize trainspiler code

* add comment to sum_op

* add warning log

* fix comment

* fix test_send_recv

* fix test_send_recv

* fix train with no distributed table

* optimize GetDims

* add prefetch_op

* fix ci

* optimize code

* optimize code

* fix include

Fix sparse update memory error for distributed training

* dynamic send/recv selected rows

* update by comment

* fix by comment

* initialize batch barrier

* add some comments

* update

* fix batch barrier

* use sendvariable rpc interface to send batch barrier

* fix comment

* fix method

* fix by comment

* fix by comment

* Updating the cluster trainign doc

* Fixed comments

* Updating few comments in recv_op