A fix for a recent bug in harvesting from some generic OAI archives

[landreev]

What this PR does / why we need it:

This fixes an annoying harvesting bug that crept into 6.6. As we were focusing on some new harvesting features being added, such as being able to harvest from Datacite and some other archives with unique and idiosyncratic features, something very basic got broken - namely being able to harvest from the most basic/generic non-Dataverse OAI archives (for example, DSpace - as was reported in the users group in recent week).

The fix is extremely straightforward - 5 or so lines of code that needed to be dropped.

It is quite late in the release cycle, but because the fix is so trivial - and also quite useful to many users, apparently - I strongly suggest that we add it to 6.7.

Which issue(s) this PR closes:

• Closes Record import from DSpace via OAI fails #11479

Special notes for your reviewer:

Suggestions on how to test this:

The reproduction scenario submitted by the user in the issue above can be used to verify the fix (harvesting from a DSpace OAI archive):

Create a harvesting client with the following settings:

oai server: https://darchive.mblwhoilibrary.org/server/oai/request
oai set: col_1912_4841
metadata format: oai_dc
archive type: "Generic OAI archive"

Harvesting will fail in develop branch.
Should work, and harvest 7 records with this build. (It may be necessary to erase the harvesting client and re-create it from scratch before retesting).

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

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

Additional documentation:

[github-actions]

📦 Pushed preview images as

ghcr.io/gdcc/dataverse:11479-harvester-broken-generic-oai

ghcr.io/gdcc/configbaker:11479-harvester-broken-generic-oai

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

[ofahimIQSS]

Before:

After:

PR validated - no issues to report.

@landreev It looks like continuous integration is failing (ansible)

[landreev]

I never got around to looking into the continuous integration issue last week, but on it now.

[landreev]

P.S. Actually, the last 3 failures are all 4 sec. "ansible job terminated abnormally" - i.e., seem like Jenkins flukes rather than actual test failures. Let me re-run and see what happens.

[landreev]

Once the tests actually ran, everything has passed apparently.

[ofahimIQSS]

Awesome! Merging.