Put it in top gear and push it along the floor [with key OFF] to make sure engine is OK. That bundle of earths should all go to the rear engine mounting. Make sure they are all clean and tight.
When the starter first stopped on compression, I put it in 2nd or 3rd and pushed it over the compression stroke, thinking that if I got to the other side. momentum would help push it over. That didn't work, but since I'm not over-heating, and everything else is good, I have no reason to suspect the engine.
With the battery removed, the starter still had some trouble going past compression, but not having much experience with big twins, I'm not sure how much of it is normal.
I pulled the harness apart. The good news is that I only have to replace about six 18" lengths of wires that are in the harness, and while those ground wires are fried, they didn't take any others with them.
I have the manual, now. There's a whole chapter devoted to the starter, alone. It even calls out measurements on Fluke meters, of which I own. Reading through the manuals, it will say things like "replace the brushes", as a example, but then there's no part numbers anywhere (I can find) for the brushes. Still, the manual has very detailed test procedures for the starter, so I'll be able to find out exactly what part of the starter is failing, if any. McMaster-Carr may have brushes, if it comes to that.
One thing that is missing, though, is a wiring diagram of all the wires color-coded. I've got something like 5 black wires and one purple one that was in that collection of grounds. A purple ground? In many cases purple is a hot wire color. I know it's a ground, so I'll find some purple wire (and black wire) and repair the damage. But I do want to know where all these wires go. Why did they fry? I need to find that answer. I still suspect I fried the trigger side of the solenoid. But again, if that happened, a fuse did not do its job.
The main ground cable goes to a bolt that goes through the frame and into the engine. This is very good because it grounds the frame and engine with one bolt. Thing is, now I'm working my way to that bolt so I can make sure it's not corroded, even though the meter says continuity is good.
Next, I need to go through the fuses. A fuse was not supposed to allow this to happen in the first place. This is an interesting problem, because the main ground cable taking the high voltage to the engine case appears to be fine, and tests good.
I got a kid graduating today. Maybe I can slip out and tinker with it some more.