Take some scrap cable material. Grab it with your hands a foot or two apart. Run it over the edge of a board or the edge of your bench with some tension in it. Everywhere it touched the edge will now be curly. When they serve the cable strands and then stretch it they are putting very slight yield and memory into the cable. That is why the strands stay together and it behaves like a larger wire, but is more flexible. When you do anything that re-stresses or reforms the cable you unseat the memory of it being twisted into a cable. There are lots of ways to do this.
You can start getting squiggles if you put some twists into the pair of lines by flying some loops, then try to brute force the twists when you run the lines out next time by either putting them between your fingers or by squeezing with a rag and pulling toward the handle. Do you keep pulling until the handle is hanging in the air and waiting as it untwists? Not always curlies, so there may be other things needed to cause that kind of damage. But certainly squiggles and kinks. As one of our southern hemispherical brethren mentioned, letting the lines come off the spool in loops (ie. not rolling them on or rolling them off; as in don't wind them like kite string!) will either put a lot more twists into a cable--or will take them out.
Combat line get curlies all the time when they cross lines and the circular cable cross sections are upset. Think point-loading again. Edge of bench is just someone else's lines, now.
I always figured that the reason that early in control line modeling, guys tried to use swivel line connectors, was to take care of individual wire twist. It would help neutralize some of the ways cables can be damaged, but not most ways and not the most likely ways. However, swivels were too weak and unreliable and caused way more serious problems than the minor line-handling issues that they helped... and were correctly outlawed in my opinion.
I always understood that the cable was effectively stretched during manufacture. But I have often taken a new set of lines and hooked them up to the fence and pull tested/stretched them before use. If they pop I'd rather they didn't do it when hooked up to hundreds of hours of work on balsa and paper. And, it should "seat" the strands if they weren't already tight. I use a gage because people are not well calibrated when you try to pull X-pounds. Overstressing them is probably just as bad as not testing them at all.
The toughest to handle are the .008 stranded 1/2A lines. But they fly so good ya gotta have 'em. If you can get used to handling those without damage, then all the larger sizes are easy. Solids have other handling issues....
McSlow