* fix nccl2 lars dist support

* put lars in momentum op

* add tests lars

* fix ci

* fix cpu kernel

* soft warning

* remove lars in test_recognize_digits.py

* move to another op

* add file

* update api.spec test=develop

* update test=develop

* fix api.spec test=develop

* wip

* wip, finish grad merge ops

* wip, finish graph build

* wip test running

* work on 1 gpu

* workable version

* update

* fix tests

* fuse broadcast op

* fix compile failed

* refine

* add batch merge test mnist

* fix CI test=develop

* fix build

* use independent bn params for batch merge test=develop

* update api.spec

* follow comments and for test

* wip

* refine tests test=develop

* follow comments test=develop

* remove startup bn modify test=develop

* follow comments test=develop

* fix merge test=develop

test=develop

test=develop

test=develop

* fix graph num hang test=develop

* re-enable tests test=develop

* re-enable graph num check test=develop

* fix multi device pass role check test=develop

test=develop

test=develop

test=develop

please fix offline

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

Make all blob info saved in global device context to be thread based.
Meanwhile save thread id in thread local storage in ParallelDo

test=develop

test=develop

test=develop

test=develop

1. Fix CUDA kernel when height is large than 2048.
2. Support input with more than 2D.
3. Fix unit test when k is large than 1.
4. Enhence unit testing.

test=develop

test=develop