Partner Imports: Detect low quality publishers

[cclauss]

Closes #6573
Closes #6604

Create a set of LOW_QUALITY_PUBLISHERS (is there a better name? SPAM_PUBLISHERS?) and then for each book create a set of publishers. IF there is notebook in the book's title AND there is an intersection between the two sets THEN it is a low-quality book that we should not import.

Technical

Testing

See: scripts/tests/test_partner_batch_imports.py

>>> a = {p.casefold() for p in ["razal", "tobias publishing", "koraya", "pickleball", "d"]}
>>> b = {p.casefold() for p in ["hol", "mad", "mazz", "mikemi", "tobias publishers"]}
>>> c = {p.casefold() for p in ["pickleball publishing"]}
>>> a & low_quality_publishers
{'tobias publishing'}
>>> b & low_quality_publishers
set()
>>> c & low_quality_publishers
{'pickleball publishing'}

Screenshot

Stakeholders

[RayBB]

I really like this approach!

[cdrini]

Can you remove the last commit? It looks like an autoformatter ran with it and I can't see the diff anymore

[cclauss]

Oh, Fudge! I got sick of hand wrapping long lines so I ran black. big mistake. I will do the deed in my morning.

[cclauss]

All quotes restored. Created #6624 to avoid quote grief on future pull requests until the follow-on to #6612 has landed.

[cdrini]

Ok! I believe this should do the trick. Notable changes I made:

1. These publishers are actually (confusingly) imported as authors ; so check the authors field
2. created isn't on the record; publish_date is the field we seek
3. The publisher-author exclude list is a hard exclude; it doesn't require the title to contain anything. Those publisher-authors should always be excluded

Let me know if anything looks off! If all is good, please feel free to merge :)