I ran a basic test using an LED and resistor connected to GPIO 14 — unfortunately, the LED stayed at the same brightness regardless of the spindle speed (S-value), which suggests that GPIO 14 is not outputting a PWM signal.
However, when I toggled the spindle on via the FluidNC web interface, I noticed a voltage change on GPIO 13, which indicates that the 10V: section of the YAML is most likely being loaded and interpreted correctly.
So it seems that FluidNC is parsing the config, but GPIO 14 isn’t behaving as expected.
I haven’t had a chance to set up a spare jackpot to test, and probably can’t today.
It might be worth doing a test where you remove the 0-10V module, change your config.yaml to swap GPIO 14 and GPIO 13, then repeat your test using the LED in GPIO13.
If you get the ability to change intensity of the LED when commanding spindle speed with GPIO 13, then you may have a bad GPIO14 on your ESP-32.
Bonus points if you test GPIO14 again after you have pulled the 0-10V module but before you modify the config.yaml. This rules out a weird failure on the 0-10V module. I don’t expect that to be the issue.