metadata export for drafts via API

[pdurbin]

What this PR does / why we need it:

#11305 explains that there are use cases where it is desirable to export metadata from datasets while they are still in draft. This pull request delivers this functionality, available via API.

Which issue(s) this PR closes:

• Closes Feature Request: Generate a Croissant metadata file (or any export format) before a dataset is published #11305

Special notes for your reviewer:

Drafts are exported on-the-fly rather than being cached.

The export API is a bit non standard in that it doesn't support our pattern of being able to pass in either the database id or the PID of the dataset. Only the PID is supported. I didn't try to address this. My changes are backward compatible.

From https://github.com/gdcc/dataverse-exporters I only made a pull request to update the Croissant exporter. Once we merge this PR perhaps we can create issues for the remaining exporters to update them as well. Also, in that repo I believe we need to add some more docs to explain that if you upgrade to Dataverse 6.7 you should update exporter whatever to version whatever so that drafts are supported. I gave a heads up about this in the release note snippet.

I had to edit src/main/java/edu/harvard/iq/dataverse/harvest/server/xoai/DataverseXoaiItemRepository.java. I'm not sure the best way to test it.

I took a quick look at making the "export drafts" functionality available via UI but there are a few challenges:

• We're trying to touch JSF as little as possible with the React UI on the horizon.
• JSF constructs a URL that works well for published datasets. No API token is included. For drafts, I didn't want to introduce a security risk by simply adding the API token to the URL. I looked briefly at the "session user" concept in the Data Access API but it seems non-trivial to support it.
• There's a fair amount of logic in the dataset page for exporting: (!DatasetPage.dataset.deaccessioned or (DatasetPage.workingVersion.deaccessioned and DatasetPage.canUpdateDataset())) and !DatasetPage.anonymizedAccess. I didn't want to break anything. By the way, the file page logic is simple (FilePage.fileMetadata.datasetVersion.dataset.released) but perhaps it should match the dataset page? 🤷

Suggestions on how to test this:

• Create a draft dataset.
• Follow the updated API docs and download the draft
• Test all builtin exporters but note that exporters that rely on ddi (ddi and html and) and schema.org (schema.org and croissant at least) had to be updated.
• Export drafts from the Croissant exporter which you'll have to build yourself from handle drafts gdcc/exporter-croissant#14

Does this PR introduce a user interface change? If mockups are available, please link/include them here:

No.

Is there a release notes update needed for this change?:

Yes, included.

Additional documentation:

Preview docs at https://dataverse-guide--11398.org.readthedocs.build/en/11398/api/native-api.html#export-metadata-of-a-dataset-in-various-formats

[coveralls]

coverage: 22.997% (-0.009%) from 23.006%
when pulling 23c990d on 11305-export-drafts
into e3bc7cf on develop.

[github-actions]

📦 Pushed preview images as

ghcr.io/gdcc/dataverse:11305-export-drafts

ghcr.io/gdcc/configbaker:11305-export-drafts

🚢 See on GHCR. Use by referencing with full name as printed above, mind the registry name.