Glad I'm not the only one, because that's exactly where I'm at. The premises almost relies on consistent yield and unconstrained growth, which nature very much does not like. Plus it doesn't consider the opportunity cost of having to sink your time into becoming a literal farmer (nor any other recurring costs to maintain and harvest your plants).
In this case, the upfront cash is hands down the way to go. You don't even have to do any complicated investing, just huck the mill in a jumbo CD and take the monthly payout. Going off my local credit unions (about 3.75% in dividends for a 5 year term), at $37,500 per year it probably wouldn't be enough to quit your job, but you'd be doing an order of magnitude better than $50 per week. If you're really looking to grow it, you could just dump the lump sum in the S&P 500 (up 95.3% from 5 years ago). (Assuming no taxes and that the dollar still has any value in the next 5 years).
I mainly use my printer as a tool to solve problems, so my decision process is very much grounded in arriving at a solution as opposed to just finding something to keep the machine busy.
My usual approach is to cast a wide net and go through all the models that might do what I'm looking for. If I'm lucky, I'll find something that I like enough to print. If not I'll use it as a brainstorming session and either pick out a model or two that I can adapt (or at the very least pull critical dimensions) or get a feel for what I'd like to do differently. From there, it's off to CAD where I'll fire off slivers and prototypes until I'm happy with the fit and function of my part.
printables is my go-to, but sometimes I'll wander over to thingiverse if I want more options (and know I'm not working on something bespoke)