Aurora inference notebook on JASMIN (PET)

Aurora is a foundation model that has been shown to be stable on climate time scales
- [aurora info](https://www.microsoft.com/en-us/research/project/aurora-forecasting/)
- [more model info](https://machine-learning-made-simple.medium.com/understanding-how-microsoft-ai-built-a-foundation-model-for-climate-forecasting-679e81fffc2d)
- [github repo](https://github.com/microsoft/aurora)
- [docs](https://microsoft.github.io/aurora/intro.html)

## Subtasks

- [ ] create environment on JASMIN
- [ ] download model
- [ ] prepare initial condition data
- [ ] prepare demonstrator tutorial
- [ ] add Aurora to PET
- [ ] demonstrate using Aurora in PET pipeline
- [ ] run for extended period, calculate climatology
- [ ] run with downscaler?

No project files or tests are named. Start with the linked Aurora repository and documentation, then determine how PET is structured before creating the JASMIN environment and preparing the initial-condition data. Done includes a demonstrator tutorial, PET integration, a pipeline run, an extended run with climatology, and an assessment of downscaling.

Written by the indexing model from the issue text.