feat: add depth slider and cache figures in BNN prior demo

feat: add SLURM array job for BNN prior depth sweep

fix: update save directory default and optimize inference for CPU device

refactor: inline inference into main(), add CLI args and save all artifacts to parameterized dir

fix: update bnn-priors git source from SSH to HTTPS in requirements and lock files

fix: update SLURM array range and reduce job time; add steps parameter to script

feat: work-in-progress on prior demo

feat: add SLURM array job for grokking sweep

fix: config adam correctly

fix: include pyarrow

revert: restore PEP 723 script for grokking training

feat: switch to CUDA 12.1 for PyTorch

feat: add CUDA 12.1 support for grokking script

cleanup: remove compute workspace and adjust layout

refactor: convert grokking script to PEP 723 format

feat: add grokking training script for modular arithmetic

feat: add compute workspace for heavy model training