This is an interesting observation. Without knowing more details, I suspect the answer could be:
1. The gas-saturated semi-crystalline polymer forms bubbles because gas solubility generally decreases with increasing temperature. This is the same thing that generates air bubbles (not steam) on the sides of a pot when you heat cold water.
2. Thermally induced crystallization is widely documented for many systems and can results from several phenomena, including (a) thermal energy helping to overcome energy barriers to systems that want to crystallize but are in a meta-stable, non-crystallized state, (b) bubbles forming nucleation points for crystal formation, and (c) chemical changes. I suspect there are other mechanisms and I suspect that your particular case is either (a) or (b).
Any system with optical access would do this. I am told (I don't drink soda) that a bottle of soda that has been placed into and then removed from a freezer and opened quite commonly does this - that is - gas bubbles come out of solution while the remaining liquid forms slush (crystals suspended in liquid). I am not sure how long it must spend in the freezer, but I am quite sure this is an important variable for this to work.