[Do not merge] Speed up LB tests

[amCap1712]

This is a catch all changes PR to benchmark speeding up LB tests. It touches lots of parts of the codebase, some with partial or no test coverage. As such its not possible to review it well, the intent is to keep this PR as a record of all the changes that need to be made and open smaller PRs and improve speed incrementally.

[pep8speaks]

Hello @amCap1712! Thanks for opening this PR. We checked the lines you've touched for PEP 8 issues, and found:

• In the file listenbrainz/db/listens_importer.py:

Line 43:20: W291 trailing whitespace
Line 73:54: W292 no newline at end of file

• In the file listenbrainz/db/recommendations_cf_recording_feedback.py:

Line 49:131: E501 line too long (135 > 130 characters)

• In the file listenbrainz/db/similar_users.py:

Line 133:49: W291 trailing whitespace

• In the file listenbrainz/db/tests/test_feedback.py:

Line 106:29: E128 continuation line under-indented for visual indent

• In the file listenbrainz/db/user.py:

Line 435:45: W291 trailing whitespace
Line 439:33: W291 trailing whitespace

• In the file listenbrainz/db/user_setting.py:

Line 66:15: W291 trailing whitespace

• In the file listenbrainz/tests/integration/test_api.py:

Line 20:1: W293 blank line contains whitespace
Line 621:1: W293 blank line contains whitespace

• In the file listenbrainz/webserver/views/test/test_player.py:

Line 18:37: E203 whitespace before ':'

[MonkeyDo]

As an aside, I recently removed the action-docker-layer-caching steps in the BB repo github actions after seeing reports of improved speed, and indeed my builds are now a lot faster without the caching steps.
Maybe something to remove from LB too?

[amCap1712]

Oh, I see. The caching does take a lot of time so makes sense to remove that if the actual build time is less. I'll do some quick tests and will remove if its slower. Thanks!