You are mistaken. These reactors did not produce any excess heat. If they had, there would not have been a 1-year test. They would have paid Rossi, and they would be developing the gadget into a commercial device. They had a brilliant staff and tons of money lined up to do that.
I think it makes more sense to first make a long-term endurance test and then after that pay the 100 millions and develop the commercial device.
That's probably why the 3rd payment was the largest chuck, it's not enough to have a working technology, it also needs to be safe and commercially viable. The IP might not have been worth 100 millions if the technology was not commercially viable, but by testing +50 reactors for a whole year they would derive 50 years worth of operating data. It's just due diligence.