v0.3.7
[shardformer, pipeline] add `gradient_checkpointing_ratio` and heterogenous shard policy for llama (#5508)
[moe] fix mixtral optim checkpoint (#5344)
[doc] fix ColossalMoE readme (#5599)
[devops] remove post commit ci (#5566)
[moe] init mixtral impl
[fix] fix typo s/muiti-node /multi-node etc. (#5448)