[x] Update all redirectors to root://xcache/ to make sure we are hitting the nearby storage
[x] Get rid of distributed notebook, do not expect to make such a distinction with the dask tutorial upstream
[x] distributed correctionlib example with lumimask p
[ ] ~Should update to scikit-hep/vector anywhere appropriate, to be consistent and future-proof for EOL coffea.vector~: scikit-hep/vector still has some integration work to do, we'll stick with tried-and-true for this moment
[x] ~Add example of using dataset discovery tools to find datasets and add metadata before preprocessing them~
[ ] Optional: Add example of using mltools and Triton IaaS (can be based upon untrained ParticleNet, just to show that it works...)
root://xcache/to make sure we are hitting the nearby storage