I have done extensive testing for hydrogen embrittlement/de-embrittlement in the past, and unfortunately I have never come across a non-destructive method, or for that matter, a fast method (at least, not beyond hitting the screws with a hammer level of testing). To make matters even worse, there is nobody (that I have found) who will state that any given level of hydrogen contamination of a part is 'bad'. They all agree that the more there is, the worse it is, but there is no 'cutoff' point.
As was pointed out earlier, the method of removing hydrogen embrittlement is well understood, and my suggestion would be to load the requirement onto your supplier; have them provide evidence that each lot has been properly processed to relieve hydrogen embrittlement.
The torque testing noted earlier would also be a good method, and would allow you to independently verify your suppliers findings.