If some backend fails, the failed proof step is reported to the IDE, listing all the assumptions and with the symbols expanded. But then, when the next backend is attempted, the proof state is reset to "proving" with no details on the step. The details could be retained until the step is proven to make the information visible to the user.
This is related to #93.
If some backend fails, the failed proof step is reported to the IDE, listing all the assumptions and with the symbols expanded. But then, when the next backend is attempted, the proof state is reset to "proving" with no details on the step. The details could be retained until the step is proven to make the information visible to the user.