calc_WodaFuchs2008() deserves more attention

[mcol]

From #197:

• The curvature will be infinite if the 1st and 3rd bin have the same count (which is not a remote possibility) as it divides by counts_log[1] - counts_log[3]
• It's unclear why we should only use the first 3 elements of count_log, as if the user sets a large number of histogram breaks, then count_log will be much longer than 3
• At the end of the day, it doesn't quite matter as curvature is never used anywhere in the code after it's computed!
• There is a minimum number of breaks set at 4 which seems somewhat arbitrary, in particular if the curvature computation is unused

Moreover:

• Should the function add the data slot to the RLum.Results object it produces?

[RLumSK]

@mcol We should have a look together after the 0.9.25 release.