:sparkles: Enable regression & misc

[x] Compress GNNL and NNL
[x] Make the MLP more flexible
[x] Do we really need to add a parameter to the CLI (regression vs. classification)? Why not use dm for this?
[x] Add VGGs
[x] Add LeNet
[x] Add a setup.py when poetry is not available + mention it on the installation page
[x] Improve the MLP baseline
[x] Add ensemble regression
[x] How to deal with mutlivariate regression cases? (different output) -> Ask for the dimension first
[x] Fix Bayesian layers? (cf failed tests: #28)
[x] Add bayesian models & losses
[x] Add deep ensembles method
[x] Add a Bayesian tutorial
[ ] Fix tutorial generation
[ ] Make tutorial gallery static
[ ] Add a pretrained tutorial
[ ] Add a regression tutorial
[ ] Add tests
[ ] Add docs
[ ] Improve VGG optim recipes?

Done for now: Add an arg to CLI to split regression and cls :hammer: Modify tests accordingly :heavy_check_mark: Add a regression recipe :sparkles: Add a small MLP model :sparkles: Add an MLP baseline :sparkles: Add a UCIRegression Datamodule :sparkles: Add GaussianNLL lightning metric :sparkles: Add an experiment :sparkles:

ENSTA-U2IS-AI / torch-uncertainty

:sparkles: Enable regression & misc #27