Duplicates workflow

[Webreaper]

Now that the hashes are created (see #78), need to thrash out the workflow for managing duplicates. High-level:

1. Duplicate page lists images that have duplicates (by image data). Show one image on the left, and a list of duplicates thumbnails that match on the right.
2. Checkbox to include "visually similar" duplicates
3. For each duplicate on the right, provide a checkbox to 'action' the duplicate. Note that once n-1 checkboxes are checked, the remaining checkbox (which is deemed the 'original / to keep' image) should disable itself.
4. Checked duplicates can then be actioned by a button at the bottom of the screen. Options for processing are:
   • Delete the duplicate images (BIG warning)
   • Move to a new folder (generic duplicates folder? Or let the user choose?)
   • Images tagged with keyword (default: duplicate)
5. Perhaps have a bulk mode to auto-select the right images. For example: "Keep oldest", which will check the checkboxes of all but the most recent (date taken, or date modified).
6. If the auto-selection of the images to keep is incorrect, the user could switch them by drag/dropping another image from the right-hand-side onto the main thumbnail.

Will look something like this. Control can also be used for person/face tagging workflow, per #227

Anything else?

[irrelevantdotcom]

Will it be possible to leave duplicates in-place, but have a link to 'other instances of this image'? Reason being I'm about to embark on digitising my late dad's extensive photo archives (right back to 1950s, it's daunting) and I'll be scanning both the negatives, and prints. It would be nice to be able to tie these together, since they will be coming in at different times and within different folders, depending on where and when they were used.