Taxonomiees: Improve Tags/Categories response size by limiting the requested fields

[youknowriad]

closes #3913

Request only the required data from the taxonomies endpoints.

Testing instructions

• Create a post
• Create a tag
• Type in the tags input
• Check that the triggered requests only return idand name for each tag
• Assign tags and check that it saves properly.

[jorgefilipecosta]

Tests well for me 👍

[adamsilverstein]

Great!

[aduth]

Noting that withAPIData has built-in options for caching requests, so varying the ?_fields parameter would be treated as a separate request.

[adamsilverstein]

@aduth that should be fine here though, right? since the ?_fields won't change? i'm not sure these requests should be cached because terms could change on the server and you would have to flush the cache (reload?) to pick up changes?

[aduth]

I guess in this case there's no difference since the hierarchical and non-hierarchical term selectors are querying different data anyways (categories vs. tags), but in other cases we're reusing the same data across many components so filtering fields could cause multiple requests unintentionally.

i'm not sure these requests should be cached because terms could change on the server and you would have to flush the cache (reload?) to pick up changes?

...Something something hard things in computer science 😄

To the previous point, however, we definitely don't want to be firing 20 requests for the same endpoint just because 20 components happen to be using the data. Some amount of caching is quite reasonable to expect.

[adamsilverstein]

Some amount of caching is quite reasonable to expect.

Absolutely! The caching you describe sounds ideal. Also, thanks for the link to the hard things source, I had only seen some of the riffs.

In this case the request is an admin side search so minimal/no caching seems appropriate.