Just get it as good as is reasonable. I only know one sure home measurement method, and it needs the fork off the frame, on a flat table ideally a reference plane, some precision tools, parallel blocks, et cetera. Find a method you can sight in with reasonable ease, do it about 5 times recording the numbers prob to the millimeter each time, and calculate the average. Or post all the numbers and I'll do the math, it's falling off a log for me (engineer with 30 years). Then I'll get you a trail value that's as good as the inputs are, look it up in my Talbot, and see what he said about that range, for what that's worth.

For a built bike, I fix the bike in a training stand and level it as well as possible. I align a meter stick with the stem center bolt and the center of the fork crown, and make sure the bottom of the stick is sitting on the floor. Secure the stick to the bike with tape. If necessary I mark key points on the bike with a fountain pen, which just wipes off of painted surfaces. Then measure the offset with a metric scale, perpendicular to the meter stick edge (which now represents the steering axis), to the center of the quick release skewer. That's the offset. You can see the setup is tricky. Sometimes I can reproduce the manufacturer's numbers.

I'm gonna guess it has a lot of trail. Short wheelbase usually means a front wheelthat's kind of close to the downtube, which implies not much offset. A small value of offset will compute out as lots of trail.

I really have a similar "re-create the past" project, my old mid-60s Rossignoli road bike. I think it was a great example of the race bike for not-so-good Italian roads of the day. I do wish I still had it (stolen!!!), or at least knew the numbers so I could work with a framebuilder to see if I should re-create it.

For the rationalists here, we are clearly beyond the pale.