Interesting.
I used to be a metallurgist, & worked for a while in a tube works.
A quick look under scope would soon tell.
For work hardening leading to failure you have to plastically deform metal, ie bend it, not just spring it. Thermal expansion /contraction works out to be very little movement. This pipe doesn't look to have fractures in the right way to fit that. It's actually very hard to stop a pipe moving the sort of distances involved so you only get a problem with a long length, and it wouldn't cause this longitudinal cracking, it basically squashes the pipe at bends and tees, and makes solder joints fail.
This looks a bit like freezing damage, but you always get expansion of the diameter when that happens, even in the unsplit parts. In my experience it goes half a mm or so before something gives. And the part where the ice pushes through the copper, it really pushes the pipe out locally, because it weakens it as it goes thinner. Any sign of that?
Looking again at your pictures the pipes are deformed - is all of that where you hauled it out of the floor?
Corrosion is a main contributor, would be my guess, without looking closer. If you get bad pipe it'll have sulphur in the copper, which makes it brittle and also allows intergranular corrosion. The grains are extruded out during manufacture into linear "fibering" so the lines of corrosion go generally along the length. Any mechanical deformation would make it worse.
There has been a lot of crappy pipe, I've come across it in housefuls. Imported during shortages, things like that.
Sometimes you get a garden-hose effect, with a line of little holes corroded through, mostly in lines along the pipe.
Copper doesn't tend to corrode a lot of course, until someone leaves flux residues or similar in there with chlorides and acidity. The you DO get the intergranular corrosion. That sort doesn't show much on a surface, it goes initially for grain boundaries, then causes cracks. In the cracks you get little electrolytic cells doing their thing and making it accelerate.
The next thing you get , is movement from heating and cooling which is in the elastic region of the stress/strain curve. If you get huge numbers ( millions) of cycles, wings drop off aeroplanes, and that, and it's called metal fatigue. It's a vibration thing really.
You will not get enough cycles to cause metal fatigue itself, but there is a similar effect with fewer cycles, down at fine level, where new corrosion sites are exposed as the movement occurs. It's called "Stress-corrosion cracking".
The characterisic fracture surface is not deformed, but a bit faceted. You don't see gross corrosion like you would on the surface of the pipe, unless it's been an aggressive chemical. Stress Corrosion Cracking is much slower and smaller scale.
Does that fit?
That deformation - if the pipes were weakened by SCC, then increased pressure from freezing could easily make it "go" in long sections, I'd have thought.
Done just the one job with it , as an apprentice - often wondered how long it lasted 
