Sleepers version 2.2.0

[sanderegg]

What do these changes do?

• use latest Dockerfile best practices
• standard sleeper is again below 150MB
• sleeper-mpi/-gpu use latest Cuda 12.3.1
• adds dream input to generate a file of any size in bytes
• adds dream output as file of the dream input size for testing
• simplify generation of synchronized sleeper/-mpi/-gpu with 1 single metadata file
• upgraded Github workflow file
• version bumped to 2.2.0
• uses latest uv version in Github CI

details

• sleeper uses:
  • python 3.11-slim image
• sleeper-mpi/-gpu uses
  • nvidia/cuda-12.3.1-base-ubuntu-22.04 image

Related issue number

Checklist

• I think the code is well written
• I updated the table of contents (make toc)
• Unit tests for the changes exist
• Documentation reflects the changes
• Runs in the swarm
• If you design a new module, add your user to .github/CODEOWNERS

[pcrespov]

thx for this big cleanup!

[matusdrobuliak66]

Thanks for the upgrade! :) please consider putting even higher upper limit (so we can run it let say for 3 hours?) thanks!

[bisgaard-itis]

Very nice! Thanks a lot

[elisabettai]

Thanks for the deep (sleep) refactoring.

@wvangeit might like it a bit more, since it the code does something (i.e. producing a dream) instead of just waiting for time to pass. 🙃

[wvangeit]

Thanks for the deep (sleep) refactoring.

@wvangeit might like it a bit more, since it the code does something (i.e. producing a dream) instead of just waiting for time to pass. 🙃

Can we let chatgpt generate some hallucinations for the dreams?