Strong Semilattice of Semigroups

[tomcontileslie]

This PR supersedes #673; I've started a new one to squash the current log of ~30 commits and have easier write access to the PR (I was previously submitting PRs to Murray's branch).

This PR is joint work with @james-d-mitchell, @MTWhyte and @reiniscirpons. It creates a new StrongSemilatticeOfSemigroups object initialised from a list of semigroups, a (meet-semilattice) digraph and a list of homomorphisms. Currently element enumeration, element comparison and multiplication is possible, which makes this a fully working Semigroups object, and it will be ready to merge as soon as it passes tests and review.

In future I will also be working off this branch to create additional features such as decomposition of certain types of completely regular semigroups into StrongSemilatticeOfSemigroups objects.

[tomcontileslie]

Just realised that there is inconsistency in the documentation calling an operation UnderlyingStrongSemilatticeOfSemigroups when it is actually implemented as UnderlyingSemilatticeOfSemigroups. I'll fix this in the next few days.

[codecov-io]

Codecov Report

❗ No coverage uploaded for pull request base (master@664af4a). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master     #722   +/-   ##
=========================================
  Coverage          ?   96.53%           
=========================================
  Files             ?      160           
  Lines             ?    37792           
  Branches          ?        0           
=========================================
  Hits              ?    36482           
  Misses            ?     1310           
  Partials          ?        0           

[james-d-mitchell]

I'm happy with this apart from some minor comments. One other comment is: have you checked that the documentation looks ok when viewed as text only?

[james-d-mitchell]

I think I'm ready to merge this now, any reason not to @tomcontileslie ?

[tomcontileslie]

@james-d-mitchell I've just updated the tests since your review updated error messages printed to the console in the standard test suite. All clear for merge on my end. To answer your question on documentation, all seems to render fine when read from within the GAP interface.

[wilfwilson]

When this PR is ready to be merged, I suggest doing a squash-and-merge to tidy up the commit history (see #752 (comment)). Or, if @tomcontileslie is confident and knowledgeable enough with git to rewrite the history of this branch via an interactive rebase and then force push a cleaned up version, then go ahead.

[tomcontileslie]

@wilfwilson I've tidied up the history into three commits that show broad chronological progression (and each commit should pass tests). Let me know whether that works for you, I can always squash further.

[wilfwilson]

Nice one!