Shouldn't red use a 4.87ohm limiting resistor if blue and green are using 1.75ohm resistors?

Details:

In the 3W RGB common anode LED the forward voltage is 2.5V for red and 3.6V for green and blue (according to the datasheet).

The forward current is 350mA for all three colors.

On the schematic for the Prop-Maker FeatherWing you can see that the limiting resistor is 3ohm for red and 1.75ohm for green and blue.

Given the standard limiting resistor equation R = (Vs - Vf) / i we can calculate the supply voltage implied by these values.

So rearranging to solve for Vs we get Vs = iR + Vf and this results in:

- Red = 0.35 * 3 + 2.5 = 3.6V
- Green/blue = 0.35 * 1.75 + 3.6 = 4.2V

I was expecting the implied supply voltage to be the same for both cases - and I was expecting it to be about 3.7V, i.e the nominal voltage of a single cell LiPo, or 4.2V (to take into account the initial fully charged voltage).

So if 1.75ohm has been chosen for green/blue in order to handle 4.2V, shouldn't the limiting resistor for red also be for the same supply voltage?

I.e. (4.2 - 2.5) / 0.35 = 4.86ohm, e.g. something like the Yageo 1206 4.87ohm resistor.

If I look at this forum answer by Rick from Adafruit Support as to what limiting resistors you should use when the supply voltage is 5V the math all works out as expected:

On red, use a 7.5ohm 1W resistor

On green, use a 4ohm 1/2W resistor

On blue, use a 4ohm 1/2W resistor

If I plug these values into Vs = iR + Vf then I get the same voltage, i.e. 5V, for all three colors. So here we get a consistent implied voltage, so why the apparent inconsistency in the Prop-Maker FeatherWing design?