I disagree. Ants often have features that are important for ID and much smaller than 1% of the whole organism. The mandibular teeth for example. Another example is the density of the pilosity (the non-erect hair covering some body surfaces). Expecting every uploader to know which parts to crop out exactly is unreasonable. I often don’t know whether identifiers will be more interested in the hair on the antennae/scapus or on the gaster. Look at the images in this observation as an example: https://www.inaturalist.org/observations/392721212 . Neither of these images have any room to sensibly crop by any non-trivial amount and if you check the file inaturalist even tells you that they both had 6000x4000 pixels (so about 6 times as large), which makes a clear and noticeable difference when looking at pilosity.
Edit: spiders and ants are probably not even special in this regard among arthropods