Support Muon + Dion optimizers

[akoumpa]

Is your feature request related to a problem? Please describe. Add support for the two optimizers

Describe the solution you'd like Use implementations available in https://github.com/microsoft/dion/tree/main or https://github.com/NVIDIA-NeMo/Emerging-Optimizers

Describe alternatives you've considered N/A

Additional context N/A

[euronymous-aithal]

@akoumpa is there a reason you dont want to use Muon from: https://github.com/NVIDIA-NeMo/Emerging-Optimizers

[akoumpa]

@euronymous-aithal thank you, I was not aware it was OSS. I need to confirm support for DTensor, but updated description.