feat: allow direct Hessian retrieval

[iamgroot42]

Hey,

Greate codebase- well-written and nicely documented! I spent the last few days looking for good and efficient ways to compute the Hessian of a model in Pytorch but almost all answers are either outdated, do not work, or are too slow (straightforward double-loops). However with a few simple modifications, it can be easily computed using this codebase:

module = AutogradInfluenceModule(
  model=model,
  objective=MyObjective(),  
  train_loader=loader,
  test_loader=None,
  device=device,
  damp=0,
  store_as_hessian=True
)
H = module.get_hessian()

I figured it would be nice to have it as part of the repo itself! I can also add a short barebones example in the examples folder if that would be of value?