Open
Description
What are you proposing (and why)?
This is a tracking issue for enabling parallel testing of formulae dependents on CI. Sometimes, formulae have 100s of dependencies (for example, certain formulae on Linux). Since each dependent test is somewhat independent of others, we can split up dependent tests across multiple runners to reduce the time-to-merge or time until tests are done (and perhaps spot failures earlier in some cases).
CC @carlocab who brought this up.