v1.8.0

• Fix attention batching issue (thanks @leonardledy!).
• Add support for XPU devices (thanks @rwood-97!).
• Check model checkpoint hashes (thanks @isaaccorley!).
• Remove unused modulation heads in the decoder (thanks @leonardledy!)
• Make activation checkpointing configurable.
• Remove bf16_mode due to severe numerical issues and bring back autocast.
• Fix dropout bug.