- It provides a unified style to express graph and kernel for user.
- It provides a unified IR to represent graph and kernel for developer.
- It breaks the boundary between graph and kernel.
- It provides more opportunities to do compile optimization.

1.Support Broadcast op
2.Support communication op as graph output
3.Optimize Communication op memory alocation
4.support hccl multi-group

Signed-off-by: leonwanghui <leon.wanghui@huawei.com>