With a $20 CAN transceiver, documentation and/or config files from the manufacturer, and a bit of Python or something, you could absolutely bench test those electronic injectors. You might even be able to pick your injection events and adjust the metering, supporting the equipment as it ages. I'd love to see Ursa Ag put in a Megasquirt engine controller [1] or Proteus [2] or similar. You can run TunerStudio on a Raspberry Pi and show it on a touchscreen on the dash.

It's possible to build user-friendly, inexpensive and open engine and vehicle controls. You don't need to have zero electronics to not have locked-down proprietary electronics, you just need to build the electronics in the right way.

[1] https://diyautotune.com/products/ms3357-c?_pos=2&_fid=69f494...

[2] https://rusefi.com/index.html#proteus

If a tractor with a clean-burning, efficient $7500k engine could be purchased and were designed around the theory that, in 20 years or so, the owner could reasonably quickly replace the entire engine (with a first-party or aftermarket solution), would that be a good solution?

The common tech that has solved these problems nicely (IMO) is network transceivers: SFP and similar modules are built according to multi-source agreements. They contain all kinds of exotic tech, and they are not intended to be serviced at all, but (unless your switch or NIC has an utterly stupid lockout) you can pull it out and replace it with an equivalent part from a different vendor in seconds, and those parts can be unbelievably inexpensive considering what’s in them. (Single-mode bidirectional 1Gbps transceivers are $11 or less, retail, in qty 2. This is INSANE compared the the first time I lit up a 1Gbps SMF link. To be fair, this particular tech may require one to replace both ends if one fails, but if you can spare a second fiber, the fully IEEE-spec-compliant interoperable ones are even less expensive.)