Plug any gaps in testing pre-1.0

Probably mostly covered by #639, but adding a separate task here just in case before 1.0.

Maybe we can also try to add some "full launch" + automation tests (search for a place, get a route, start nav)?

Start by reviewing issue #639 and the repository's existing test setup to identify which pre-1.0 gaps remain. Define coverage for the proposed full-launch flow—search for a place, get a route, and start navigation—and consider the work done when those scenarios and relevant automation tests are covered.

Written by the indexing model from the issue text.