fix cache_mode_fn consistency

[matt-phylum]

cache_mode_fn was added to HttpCacheOptions to allow different cache modes to be used for different requests. However, cache_mode_fn was only consulted when determining if a request could be satisfied from the cache, not when determining whether to use an existing cached value or whether to cache a fresh value. The test was incorrectly passing because cache_mode_fn was setting the mode equal to the same value that it would have been otherwise.

This PR makes all code paths use cache_mode_fn consistently and fixes the cache_mode_fn test to rely on cache_mode_fn working.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (c1e8b6b) 92.58% compared to head (937e587) 92.64%.
Report is 4 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop      #65      +/-   ##
===========================================
+ Coverage    92.58%   92.64%   +0.06%     
===========================================
  Files           11       11              
  Lines         1092     1101       +9     
===========================================
+ Hits          1011     1020       +9     
  Misses          81       81              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[06chaynes]

Ah good catch, and thank you for the PR! I'll try to get this in and published this weekend.

[06chaynes]

Ignore that coverage error, it's super annoying about it. Thanks for the clippy fixes!

Edit: Ha looks like clippy was a bit overzealous with that fix I'm guessing? I can add an ignore for this case if so

[matt-phylum]

I changed the wrong instance of try_into(). My IDE does not work on this project because it tries to use multiple async implementations.