🐛Addressible LEDs are broken.

[NoahK216]

Referring to purduesigbots/pros#631

Describe the bug
The ADILed device setters appear to be broken in a few ways.

• ext_adi_led_set_pixel returns the wrong port mutex on L493. It essentially tries to zero-index a port that was already converted to zero-index, since get_ports already transforms the smart port outvalue. As a result, it tries to release a mutex on the port below what the LEDs are actually plugged into. In many cases this can permanently deadlock the user program (from past testing).
• All other setters have a strange bug where it takes calling them twice, with a delay between calls to properly update the hardware state. The minimum delays between two setter calls seems to be around 70mS.

To Reproduce
This does not work

void initialize() {
  led.set_all(0xFFFFFF);
}

This does not work either

void initialize() {
  led.set_all(0xFFFFFF);
  led.set_all(0xFFFFFF);
}

This does

void initialize() {
  led.set_all(0xFFFFFF);
  pros::delay(70);
  led.set_all(0xFFFFFF);
}

This is likely an SDK bug. Sylib gets around this by updating LEDs in a daemon loop rather than just setting once.

Expected behavior
LEDs should update the first time the setter is called.

Screenshots
N/A

Desktop (please complete the following information):

• OS: Windows
• Version 11

Additional context
None

Prerequisites

• ♻️ Refactor VDML #40

[NoahK216]

This might have been resolved in purduesigbots/pros#636