Ability to configure the number of cores

Since #1466 the main appraisal loop is parallelised. Currently (I think) this will use all available cores, but we may want to make this configurable via a `settings.toml` parameter and/or a command line flag

Start by locating the parallelised main appraisal loop introduced by #1466 and inspect how application settings are loaded from settings.toml and how command-line options are defined. Decide how the core count should be configured through the supported interface or interfaces, then verify that the appraisal loop uses the selected value instead of always using all available cores.

Written by the indexing model from the issue text.