Update zenodo dataset, setting `channel.active` to True

[tcompa]

Ref https://github.com/fractal-analytics-platform/fractal-vizarr-viewer/issues/20.

1. Review the current situation: how are we setting the active attribute of a channel?
2. Decide how we should proceed in the converters. Do we always mark all channels as "active"? Do we only do it for at most N channels? ...

[tcompa]

Review the current situation: how are we setting the active attribute of a channel?

• In the NGFF Pydantic models, active is Optional and has no value by default (https://github.com/fractal-analytics-platform/fractal-tasks-core/blob/6cb2e5119faae19f1436e60c1126495110ea7cf7/fractal_tasks_core/ngff/specs.py#L54).
• In the OmeroChannel model, active is non-optional and it defaults to True (https://github.com/fractal-analytics-platform/fractal-tasks-core/blob/6cb2e5119faae19f1436e60c1126495110ea7cf7/fractal_tasks_core/channels.py#L81). Yet another time we hit https://github.com/fractal-analytics-platform/fractal-tasks-core/issues/681.
• In the manifest, active has a default to True for both converters (standard or multiplex).

How comes we have some zarrs lying around without active=true for channels? For sure the tiny one on zenodo (https://zenodo.org/records/13305156) doesn't have active metadata.

Decide how we should proceed in the converters. Do we always mark all channels as "active"? Do we only do it for at most N channels? ...

I guess that we can continue as we do now. If the user does not specify anything in the converter arguments, all channels are marked as active.

[tcompa]

After re-running examples 01 and 02 from fractal-demos, I confirm that their .zattrs does not include the active field. TBD

[jluethi]

@lorenzocerrone Something to keep in mind in the ngio context. If we can, let's make sure channels have active=True unless there's an active decision against it. That avoids many confusions about channels not being shown e.g. in vizarr ;)