Add multi-gpu training example script

It would be great to have more examples for fine-tuning in the library!

Similar to current examples for binary-segmentaion and multi-label segmentation would be great to have Multi-GPU training example. This can be handled with PyTorch-Lightning or with pure Pytorch.

It might be a .py script or Jupyter notebook. The example should showcase how to

• fine-tune a model with pytorch-lightning (or any other fine-tuning framework, even a plain pytorch)
• compute metrics with correct gathering of results (maybe torchmetrics can be utilized)
• visualize results

In case anyone wants to work on this you are welcome! Just notify in this issue, and then ping me to review a PR when it's ready.

Fixes:

• https://github.com/qubvel-org/segmentation_models.pytorch/issues/903
• https://github.com/qubvel-org/segmentation_models.pytorch/issues/896