In the last post about the hardware, I had 3 items that weren't working as expected. I'm happy to say 1.5 of them are solved, and they ended up being soldering problems, mostly caused by the stupid QFN package of the ATMEGA32U4. Unfortunately, the TQFP package is in high demand and unavailable everywhere, so I've had to use the QFN package lately, which is a pain to solder.
Anyway, the first problem, the left SD card, is now solved after reworking my soldering. I also populated a second board and it also works, so I'm relieved about that.
The second problem is the LCD, and for that I'm halfway to a solution. The problem is that I'm using 3.3V everywhere, but a 5V LCD. A bit of a pain, and it's possible that it could have worked anyway, so it was worth a shot. I played with some different resistor configurations for the contrast and found that the LCD is indeed working, though the contrast is so low I can barely read it. I have new 3.3V LCDs coming in a couple days, so hopefully that will work. But the code is correct and the LCD does initialize and show 'hello, world' so at least we're somewhere.
Still need to figure out what's wrong with the random number generator circuit.