Having now read all the posts, I can agree it’s a very unfortunate position that you find yourself in.
I’ll agree that in some images you have provided a laitance does seem present, if so it appears very unlikely that it has been removed prior to fixing.
On the subject of priming, it would normally be visible in some way.
If it was coloured, which many are, you would clearly see a coloured residue on top of the screed.
Even if it was white or clear, you would still see a clearly defined sheen to the top of the screed denoting priming had taken place.
Even after many months, this residue is apparent.
Moisture content could also be an issue to be looked at, was it moisture tested initially?
3 months holds no guarantee that it was devoid of moisture, that would also present in a way that you are showing from your images.
Was the property water tight before the screed installed?
The areas of the floor that appear completely devoid of adhesive where tiles have been lifted (tram lines only visible) could possibly denote incorrect preparation ie. adhesive unable to bond sufficiently to the substrate.
There is guess work here until you’re able to provide all relevant information.
The trouble with voids over ufh is that it can superheat the air in the voids and can pop the tiles from the adhesive, as I understand it anyway.
Typically there’s never ONE reason an install fails, it’s an accumulation of issues that ultimately leads to a catastrophic failure, which you do appear to be on the verge of.
One tile cracking is one thing, but numerous tiles in numerous places is something completely different, especially if it’s not in a straight line thro the line of an extension for example.
That’s more easily diagnosed as a shift in between new and old building works.