fix get unique timezones

[beedub]

the previous query did not actually get distinct timezones

[cclauss]

Please run pre-commit on django_celery_beat/schedulers.py

[beedub]

@cclauss updated

[codecov[bot]]

Codecov Report

:x: Patch coverage is 75.00000% with 2 lines in your changes missing coverage. Please review. :white_check_mark: Project coverage is 87.99%. Comparing base (70d381b) to head (2c23b14). :warning: Report is 23 commits behind head on main.

Files with missing lines	Patch %	Lines
django_celery_beat/schedulers.py	75.00%	1 Missing and 1 partial :warning:

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #896      +/-   ##
==========================================
- Coverage   88.19%   87.99%   -0.20%     
==========================================
  Files          32       32              
  Lines        1008     1008              
  Branches      105      105              
==========================================
- Hits          889      887       -2     
- Misses        101      102       +1     
- Partials       18       19       +1     

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

:rocket: New features to boost your workflow:

• :snowflake: Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[auvipy]

@alirafiei75 please take a look

[beedub]

@auvipy can you suggest what additional tests you want here?

[auvipy]

The rebase didn’t work. Please make it errorless again

[beedub]

it's the same test as the one below test_timezone_offset_with_zoneinfo_object_param, except that it tests with string timezone names, which as discussed, is not possible given that timezone always returns ZoneInfo

[beedub]

if you're referring to the codecov note, then i would guess it's b/c the tests only run with sqlite. can you confirm?

if so, i don't have have the bandwidth to update the test suite to also run a build with non-sqlite (eg. postgres). i can remove the sqlite/non-sqlite conditional in that code even though it'll be less performant, but i think it'll increase the codecov by virtue of not having a non-tested branch

re: regressions, there are existing tests from the original PR that introduced this change that should cover regressions. this change is a performance improvement