Review if logger is meeting current needs

[henry-wallace-phys]

Is your feature request related to a problem? Please describe.
MaCh3 is currently quite verbose making a lot of debugging somewhat tricky particularly for new analysers.

Describe the solution you'd like

• Addition of log levels to MaCh3 (for example set printing out the xsec covariance every 1/10th of the chain to be not the lowest level)
• Consideration of what is useful/not to print out
• Optional interface to pipe MaCh3-logging into a separate log file

Additional context

• Running fits using some suggested DUNE normalisations results in O(1000) parameters. Currently very hard to parse

[github-actions]

Stale issue message