I wanted to rule out that powering the Hue PCB from 12V DC to the test point connections wasn't causing the PWM spike issue. Unfortunately while trying to test the theory I destroyed the PCB. I'm still learning how to use my oscilloscope and made the mistake of connecting the ground clip and then powering the Hue PCB with mains voltage. I do remember reading that doing this will destroy things but it slipped my mind. Fortunately my oscilloscope seems to be ok. The Hue PCB is definitely not. 

I'll be purchasing another bulb to tear apart soon to do more tests. Stay tuned.