I’m happily retired so time is what I have lots of 
. The logo is now on silkscreen where it should be and looks fine. The scaling seemed to only affect the copper layer. The pull up resistor wasn’t wired quite right so I took your latest templeaf.fzz and made a few changes: I removed two loops (where the copper connected to itself twice as they can cause problems, there should only be one path to each pad. I added the pullup resistor and changed the routing slightly to give it some room. As noted you may not need the pull up resistor, there is a weak pull up (around 100K I think) that is selectable on the arduino, if that is on you may not need a pull up resistor on the board. With open drain outputs such as the sensors have, they pull down strongly, but when they are off do nothing (pull neither up nor down) so something else either the pull up on the arduino or the pull up resistor on the board are responsible for pulling the signal high. If there is no or not enough pullup the high signal may or may not be high enough to be recognized as a high by the arduino. Worse, it can most of the time be high enough to work but sometimes not and cause intermittent errors which are very hard to find. I’d suggest putting about a 10k resistor on the board to insure you have enough pull up. I also added labels for each of the sensor rows on the silk screen mostly so I could tell which set of pads are which when checking the connections still looked right after the changes. Here is the latest board version:
Research templeaf5.fzz (36.8 KB)
edit: about the grounds: you should try and only have one ground connection from any point (such as the arduino to the board or the board to the sensors) more than one path to the same ground point can cause problems.
Peter