Save old custom session id as metadata

[sumiya-kuroda]

Description

What is this PR

• Bug fix
• Addition of a new feature
• Other

Why is this PR needed?
Photon-mosaic supports datasets that don’t follow the NeuroBlueprint format by assigning new session IDs to the sessions. However, this will overwrite the original custom IDs.

What does this PR do?
This PR updates DatasetDiscoverer._infer_metadata_keys_from_folder_names() to include the original session IDs as metadata. The corresponding metadata key is renamed to id-.

Checklist:

• The code has been tested locally
• Tests have been added to cover all new functionality
• The documentation has been updated to reflect any changes
• The code has been formatted with pre-commit

[lauraporta]

Hi @sumiya-kuroda, isn't this already the default behaviour?
E.g. this folder structure:

raw_data/
└── hippocampusA42
    └── novelEnv07
        └── recording.tif

will become:

derivatives/
├── photon-mosaic
│   ├── configs
│   │   └── config_20251113_193902.yaml
│   └── logs
│       └── snakemake_20251113_193902.log
└── sub-001_id-hippocampusA42
    └── ses-001_id-novelEnv07
        └── funcimg
            ├── recording.tif
            └── suite2p
                └── plane0
                    ├── F.npy
                    ├── Fneu.npy
                    ├── data.bin
                    ├── iscell.npy
                    ├── ops.npy
                    ├── spks.npy
                    └── stat.npy

NB: hippocampusA42 becomes sub-001_id-hippocampusA42

[sumiya-kuroda]

Hi @lauraporta,

What I implemented is to add a little more flexibility on that. For example, with this PR, ses-novelEnv07_date-20250101, which is following a custom format, becomes ses-001_id-novelEnv07_date-20250101.

[lauraporta]

Alright, I'll give this a look, tests fail as well so it's going to take a longer time