I just wanted to post the solution to my post with the reset problem but the thread was locked.
We finally managed to figure out that the problem was caused by the 3.3V voltage regulator startup time. For this specific batch of PCBs, an alternative part for the 3.3V LDO was supplied which had a much slower startup time than the originally specified IC. The specified IC had a startup ramp of 0.3ms.
The alternative supplied part had a startup ramp time of 2.64ms.
This slow ramp up time of the 3.3V rail caused the processor to not startup correctly and lock up in reset. After replacing that component with the originally specified part, all the boards worked fine.