Add more examples into the manual for StructureDescription

[hungaborhorvath]

Further, correct the old examples from #763.
Add some extra tests.
Remove InfoPerformance from tst files as per 392ad82.

[codecov-io]

Current coverage is 49.55% (diff: 100%)

Merging #969 into master will increase coverage by 0.11%

@@             master       #969   diff @@
==========================================
  Files           424        424          
  Lines        222953     222953          
  Methods        3430       3430          
  Messages          0          0          
  Branches          0          0          
==========================================
+ Hits         110234     110489   +255   
+ Misses       112719     112464   -255   
  Partials          0          0          

Powered by Codecov. Last update b63d633...5f07ed8

[olexandr-konovalov]

Thanks @hungaborhorvath - looks obviously ok to me, and tests passed. Will rebase it now.

[olexandr-konovalov]

@hungaborhorvath there are further issues with StructureDescription - I am now seeing this in make testinstall log in the master branch when GAP runs without packages:

########> Diff in /data/gap-jenkins/workspace/GAP-master-test/GAPCOPTS/64build\
/GAPTARGET/standard/label/kovacs/GAP-master-snapshot/tst/teststandard/bugfix.t\
st:1081
# Input is:
StructureDescription(testG(8,2));
# Expected output:
"((C8 x C2) : C2) : C2"
# But found:
"QD16 : C4"
########
########> Diff in /data/gap-jenkins/workspace/GAP-master-test/GAPCOPTS/64build\
/GAPTARGET/standard/label/kovacs/GAP-master-snapshot/tst/teststandard/bugfix.t\
st:1085
# Input is:
StructureDescription(testG(8,4));
# Expected output:
"((C16 x C2) : C2) : C2"
# But found:
"(C8 . D8 = C4 . (C8 x C2)) : C2"
########

[hungaborhorvath]

@alex-konovalov Thank you, I did not catch this previously. At first I thought that we should simply use the "nice" option here, because that would give stable output: "(C8 x C4) : C2" in the first case, "(C16 x C4) : C2" in the second case.

However, using the "nice" option reveals a bug somewhere. I am not entirely sure where, yet. It seems to be with nice monomorhpisms of which I have no idea about. I am debugging this, but it might take some time.

Should these tests be commented out until I figure out the problem, or shall we leave them? Or should I open an issue with this?

[hungaborhorvath]

@alex-konovalov Ok, so I have reported it in #973, and I have no idea how to resolve this. It involves that GAP (using the presentation of these groups) cannot decide if a list of subgroups is already an element of a set or not.

[olexandr-konovalov]

Thanks @hungaborhorvath - let's continue discussion in #973. I though about opening a new issue myself, but decided to wait until you'll confirm that there is no straightforward fix.