101 in memory model loading

[kkovary]

closes #101

Changelogs

This PR adds support for loading models directly into memory without saving them to a local cache. This is particularly useful for environments with limited local storage, such as serverless deployments.

Current implementation:

• A new class InMemoryModelStore is created, inheriting from ModelStore.
• The download method is overridden to load model data directly into memory.
• The load method is modified to work with the in-memory data.

An alternative approach could be to add a flag to ModelStore with something like in_memory that when set to True, use in-memory loading logic; otherwise, use current caching behavior.

• enumerate the changes of that PR.

---

Checklist:

• Was this PR discussed in an issue? It is recommended to first discuss a new feature into a GitHub issue before opening a PR.
• Add tests to cover the fixed bug(s) or the new introduced feature(s) (if appropriate).
• Update the API documentation if a new function is added, or an existing one is deleted. Eventually consider making a new tutorial for new features.
• Write concise and explanatory changelogs below.
• If possible, assign one of the following labels to the PR: feature, fix or test (or ask a maintainer to do it for you).

---

discussion related to that PR

[maclandrol]

Great feature !

Could you add some test and potentially something in the tutorial to showcase the feature ?

[maclandrol]

Also ruff is complaining.

[kkovary]

Could you add some test and potentially something in the tutorial to showcase the feature ?

yes definitely!

[kkovary]

Also ruff is complaining.

I made this commit to fix the ruff issue

[maclandrol]

I will wait for the tutorial to merge.

[kkovary]

I will wait for the tutorial to merge.

I added a short tutorial section for InMemoryModelStore

[maclandrol]

LGTM, thanks @kkovary