First of all, I don't know anything about Decalin. But shouldn't the trans boiling point be higher than the cis boiling point? It makes sense to me that a trans isomer will be more stable than a cis, resulting in a higher boiling point.
Anyways, a reflux is when you have a reaction mixture that you heat in a flask with a condenser attached (coolant flowing through the condenser). Then you heat up the material to boiling, but you don't lose any reagents because the condenser condenses the vapors and returns the liquid to the flask (I'm sure you already knew that though). My point is that you aren't refluxing unless you are evaporating reaction mixture and then condensing the vapors back into the flask. If you're heating reagents, but they aren't boiling, all you're doing is heating, and I'm willing to bet you aren't getting any thermal rearrangement.
I just tried to access your patent link, but it wouldn't let me, so I can't read the details. I think they use Decalin because it's a good solvent to use for what they want to do. (i.e., it dissolves what they want to dissolve without interfering with the reaction itself). Maybe you should look for a solvent with the same properties as decalin (is it polar? aprotic? close boiling point? is the solvent donating something in the reaction?).
