Revise A/E routines

[theHenks]

Follow CM's advice and include a drift time correction in the A/E routine.
Here, in general, two options are possible:

• Do a combined fit for the Compton bands and optimize for minimal sigma function.
• Combine all "calibrated" A/E distributions of all Compton bands and do one global max{peak_height/fwhm} optimization as in the CTC algorithm for the energy

In general, the routines have often times stability problems, so I think there is no way out of the combined fits. @fhagemann Can you have a look into this?
If not, I would take care of it and then see how far I will get here. Also @LisaSchlueter might have a look here. There are definitely already code snippets available for doing such combined fits.

[theHenks]

ToDos:

• @fhagemann opens PR with changes for combined A/E compton band fits with help from @verenaaur
  (Add combined A/E fit routine #65)
  New routines will have _combined in function names
• Compare the goodness of old to new routine by looking at the distribution of residuals for a wide variety of detectors
• Check if certain Compton bands need to be excluded from fits based on poor perfomance (e.g. bad residuals, p-values, combination etc.)
• Check that the distribution of A/E classifier is centered around 0 and has a standard deviation of 1
• Implement drift time correction routine