Episode 3, Exercise 1 proposal- Image Metadata

### Brief description
Learners open a proprietary image file (e.g. .czi) using `bioio` (as already covered in episode 2). Then we present them with a series of questions they must answer by exploring the metadata (either individually or in groups) e.g.:

- What is the shape and dtype of the image? (they should already know how to do this from episode 2)
- What does each channel represent? (e.g. what is its name / what flourophore was used?)
- What stage position was this image taken at?
- What manufacturer made the microscope used to take this image?

Exact questions would depend on the example image we use + what is accessible from its metadata. It should be possible to [access the metadata directly via `bio-io`](https://bioio-devs.github.io/bioio/OVERVIEW.html#metadata-reading) using `.metadata`.

As a follow-up, we could show the learners how to save this image as a `.tiff` file via `bioio`, and ask them to explore the metadata again. It's likely that much of the metadata will have been lost, demonstrating the importance of retaining a copy in the original file format.

### Learning objective(s)

Addresses the ['image metadata' learning objective](https://github.com/carpentries-incubator/bioimage-analysis-python/wiki/Learning-Objectives#image-metadata).

This exercise would show learners how to explore metadata via `bio-io` + the types of information that can be stored there. If we include the follow-up exercise, it would also demonstrate the importance of retaining a copy of the original file to preserve this information during analysis / processing.

### Volunteer(s)
@K-Meech

The proposal covers an Episode 3, Exercise 1 lesson on exploring image metadata with bioio, building on Episode 2. Start with the linked bioio metadata-reading documentation and the existing Episode 2 usage, then choose the example image and settle the exact questions. Done means the exercise addresses the image metadata objective and, if included, demonstrates metadata loss after saving as TIFF.

Written by the indexing model from the issue text.