Andrey wrote:

Not only is that incorrect, there's no reason for you to think that it
might be.

Well, the simple explanation is the impression of a seemingly huge
difference: I saw a big boulder cracked in halves by expanding water
that got inside through a tiny crack - the amount of water is really
tiny, and still it released enough energy to do the job. That
seemingly disproportional effect raised the question. I would've
done the calculations if I had a remote idea how to estimate energy
required to break material.

The energy required to crack an object is the integral of force over
distance (for constant force, energy is force times distance).
Although the force required to crack a rock is large, the distance over
which that force need be applied is quite tiny.

as kinetic energy instead of heat; i.e. the surroundings heat up less
than if the water had frozen in the open.

So, in other words you're saying that if the container was strong
enough there would be no expansion on freezing as the amount of
released energy won't be enough?

Since there are no perfectly rigid materials, there would always be
some amount of expansion.

You'd then get pressurized
compressed ice that will expand upon release just as gases do?

Yes.

No. Upon freezing, water releases energy; usually in the form of heat.
When it's frozen in a closed container, some of the energy is released
as kinetic energy instead of heat; i.e. the surroundings heat up less
than if the water had frozen in the open.

So a refrigerator needs less energy to freeze water in a closed
container?

Not significantly (assuming that the armor in the refrigerator prevents
the explosion from destroying the refrigerator). Most of the kinetic
energy in the pieces flying away from the bursting container eventually
gets converted to heat energy.