v0.3.8
[Feature] Distributed optimizers: Lamb, Galore, CAME and Adafactor (#5694)
[moe] support mixtral (#5309)
[shardformer] refactor embedding resize (#5603)
[misc] update pre-commit and run all files (#4752)
[gemini] fix param op hook when output is tuple (#5355)
fix some typo (#5307)