I guess you need to find a PU foam recipe to give 5-10lb/CFT density.
Ideally it would not require a massively reinforced ($$$) mold. This might be hard.
The hint about Clark using concrete molds is interesting.
Even a few psi could blow apart a mold that is small, light and cheap to build.
To get strong PU foam, you might be dealing with more than only a few psi, but it would take trial and error to find out what you needed to get the physical properties right.
Maybe concrete could work?
For example, a rough idea for a twin tip 135x40 (~54"x16"):
This would be ~800 in^2 area each side.
If the pressure was 10psi, I think you would need the mold to take 4 tons pressure.
So a concrete mold with a lid made of 3 yard of concrete (8000-10000#) would hold this pressure down. Concrete compressive strength would let the bottom half of the mold be thinner, naturally, but you would need a strong floor! Maybe you would also need a forklift.
Probably a much lighter concrete mold with rebar and I beams or really heavy angle on the perimeter to bolt it closed would be the best solution for a cheaper pressure mold.
If you can cast concrete (and weld), it might not be hugely expensive, but I am not sure if worth the trouble unless you are going to pop out hundreds of cores from the same mold. The time to engineer it and the cost might still be $$$.
Of course, molding a surfboard would mean more area, more force to resist, etc.
A custom machined metal mold could work too and be much lighter and dissipate heat faster, but cost $$$$$$. No doubt there are lots of other factors, and maybe some mistakes in my thinking.
Maybe you will find them.
Even if you could find a way to make the type of foam you need with only 0.5 psi, it would still take a pretty strong mold. What it boils down to, IMHO, is these are production techniques, rather than custom one off techniques. But maybe you can find a new kind of of foam, special recipe or new technique to make it easier!
Or for $100, get enough corecell or D-cell etc to build a few boards.