Allow Centralizer to fall back on generic GAP methods

[stertooy]

Resolves #64. Of course, if there is a reason for forcing this error, this shouldn't be merged.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 57.27%. Comparing base (604b482) to head (af20e5b).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #68      +/-   ##
==========================================
+ Coverage   57.22%   57.27%   +0.05%     
==========================================
  Files          87       87              
  Lines       13095    13095              
==========================================
+ Hits         7493     7500       +7     
+ Misses       5602     5595       -7     

Files with missing lines	Coverage Δ
gap/pcpgrp/centcon.gi	64.06% <100.00%> (+3.64%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[fingolfin]

I haven't checked yet, but: is this actually a requirement for CentralizerBySeries to work correctly? Or is perhaps the test above wrong and instead the check is supposed to check that the g are elements of the parent of G / the full group of the underlying collector?

[stertooy]

Unfortunately, it seems like this is really a requirement. In both the nilpotent and in the general case I get incorrect results for the centralizer.

gap> G := SmallGroup( 16, 8 );; # QD16, Nilpotent
gap> U := Subgroup( G, [ G.2*G.3*G.4 ] );;
gap> g := G.1*G.2*G.3*G.4;;
gap> Centralizer( U, g );
Group([  ])
gap> iso := IsomorphismPcpGroup( G );;
gap> G := Image( iso, G );;U := Image( iso, U );;g := Image( iso, g );;
gap> CentralizerPcpGroup( U, g );
Pcp-group with orders [ 2 ]
gap> CentralizerNilpotentPcpGroup( U, g );
Pcp-group with orders [ 2 ]