GitLab

* [test] refactored with the new rerun decorator

* polish test case

* [readme] polish readme

* centering image

* [test] added a decorator for address already in use error with backward compatibility

* [test] added a decorator for address already in use error with backward compatibility

Co-authored-by: github-actions <github-actions@github.com>

* fix reuse_fp16_shard

* disable test stm

* polish code

* add tensor placement policies

* polish comments

* polish comments

* update moe unit tests

* fixed bugs of assigning grad states to non leaf nodes

* use detach()