[ENH] Harmonize the channel convention parameters for edc calculations

[mberz]

The following input parameters for the EDC estimation functions

• channel_independent
• normalize
• time_shift

could be removed entirely or reduced to a single or two parameters.

Favorable would be deleting them entirely if that is possible without Lundby's or Chu's method failing.
This would make the function definitions much more compact.
If that is not possible we should opt for combining them into two parameters at max.

[mberz]

I've marked this as open discussion as long as we do not know if the favored option works sufficiently well.

[f-brinkmann]

When looking at https://github.com/pyfar/pyrato/pull/117/changes I noted that the same code appears in three different functions. If we keep the functionality, we should probably move it to a private function.