Tests for complex logic in line author feature

[GollyTicker]

The line author feature #288 contains complex logic at a few places. It would be nice to have proper testing for them to improve the sustainability of the code.

Most important places:

• time difference computations (including time-zones)
• git blame parsing
• rendering of commits to DOM based on settings
  • this can be done by directly calling the toDOM function instead of using UI-based testing
• adaptive coloring computation
• continues to work in submodules
• continues to work, when vault-root != git-repository-root

We also want to document how to use the test-vault to manually test some specific edge cases when significant changes have been made.