Added a tutorial for building the AutoEncoder with PET-supported training modules

[taimoorsohail]

Adds a new tutorial notebook,
AutoEncoder_PET_Training.ipynb, showing how to train an
autoencoder using PyEarthTools pipelines with PET/PyTorch
Lightning training utilities.

Also updates the kernel metadata for
AutoEncoder_ImprovingResults.ipynb to use the standard
python3 kernel.

Changes

• Added a self-contained PET training tutorial for an
  autoencoder workflow.
• Demonstrated:
  • Himawari data loading through PyEarthTools
  • pipeline construction and preprocessing
  • train/validation split setup
  • a compact convolutional autoencoder
  • a Lightning wrapper
  • PET PipelineLightningDataModule
  • PET Train usage
  • reconstruction plotting
• Updated notebook kernel metadata in the existing
  autoencoder tutorial.

Checklist

Getting Started

• If there is not an existing issue, raise a new issue
• In the issue, state that you are intending to work on a
  contribution. This gives everyone involved the opportunity to
  discuss the best way forward.

Docstrings

• Docstrings complete and follow Napoleon (google) style

Not applicable: this PR adds tutorial notebook content rather
than new public Python APIs.

Test Coverage

• All new code is covered by unit tests

Not applicable: this is a documentation/tutorial notebook
change. No package code is added.

Documentation

• Documentation is updated as required

This PR adds a new tutorial notebook under docs/notebooks/ tutorial/.

Final Checks

• If no generative AI was used, then tick this box

Alternatively, if generative AI was used:

• Attributed any generative AI that was used in this PR
• Included the name and version of the tool or system in
  the pull request
• Described the scope of that use

Generative AI attribution: OpenAI Codex, based on GPT-5, was
used to help draft this pull request description and
checklist text. It was also used to create the tutorial notebook.
Finally:

• Mark the PR as ready to review