Issue in image sync when the original filename differs

[patkyn]

In image service if the same image with different filename is uploaded, the new filename gets added to the alternate_filename of the image. Image sync currently does not recognise the alternate_filename. Hence, occurrence records that uses the new image filename does not get synced up.

For example, in inaturalist dr1411, in the image extension, there is a reference to an image but the image has a different original filename, hence, the image for this occurrence is not appearing: https://biocache.ala.org.au/occurrences/ce94a074-0e54-420a-80ef-71e8c9c8f2bd

koh032@aws-spark-quoll-master:/data/dmgt/dr/dr1411/current/input$ sudo unzip -p dr1411.zip media.csv | grep '63167740/original.jpeg?1583832373'
39791903,StillImage,image/jpeg,https://inaturalist-open-data.s3.amazonaws.com/photos/63167740/original.jpeg?1583832373,https://www.inaturalist.org/photos/63167740,2020-02-23T22:47:16-08:00,Carol Deane,iNaturalist,http://creativecommons.org/licenses/by-nc/4.0/,Carol Deane,63167740

Another example is related to the biocollect image mentioned in Example 2 of https://github.com/AtlasOfLivingAustralia/data-management/issues/732

[patkyn]

https://github.com/AtlasOfLivingAustralia/image-service/issues/161