For #1 and #2 does it make sense to pick a commodity first? In the current program, you don't pick a commodity first, you pick the station first, so you have a list of all stations each of which could have a different commodity with the best profit. By picking the commodity first you filter out all the stations that don't trade well with that commodity, so you have no know the trade profits to pick the right commodity instead of the right commodity being selected by the profit. Same for the end station. If you select a commodity to sell, then somehow you have to pick a station before that sells it. I would highlight the end station in the selectable stations list so that once you get within the trade range of the station, if there is a profitable trade and you see it in the list, you can select it, finishing the chain. Or you can ignore it and keep going if there are further trades. If there is no market at the destination, or no good trade, once you are within the trade range, you can just leave the end destination with no commodity exchange. You will make no profit on the last jump, but will end up at your destination.
For #3, I think my comments above apply and the code you have already actually works just fine. I think that adding stations in the middle is problematic, adding a station in between two stations with high profit is likely to have a low profit. Buying ice from the arctic and selling in the desert is high profit. If you add a middle man in the middle, you won't get a better profit by selling something to the middle man and buying something else from the middle man to sell to the desert.
It would be nice to have the option for no end station if you want to go where the trail leads you, like now.
Markup below, purely personal preference of having the distance/travel data together, the system state info together. Given I am assuming how the code works, feel free to throw suggestion in the trash!
