Feature/minimum degree topology

[mgovers]

Fixes #433

[mgovers]

Cfr. our discussions, we run the benchmark case and if the amount of fill-ins is the same for both implementations, it's ready for merge.

The amount of fill-ins differs on the benchmark case: for the meshed grid cases, all runs return 91 fill-ins for the current main and 98 fill-ins for this branch, which is a regression.

Further investigation needed

[mgovers]

cfr. discussion: boost and this branch use slightly different choices regarding the chosen point, but they scale similarly:

• sometimes, boost is better (e.g. benchmark case for large grids),
• at other times, this branch provides less fill-ins (e.g. benchmark case for small grids)
• they both are not necessarily optimal solutions

An efficient exact optimal minimum degree algorithm is proposed in https://epubs.siam.org/doi/epdf/10.1137/1.9781611976465.45 that could provide speed-ups.
Simpler solutions also exist, like randomly trying n vertices with the same minimal degree instead of only a single arbitrary vertex of minimum degree (e.g. the first, which is the one implemented in this branch)

For now, this solution seems to be OK and it is ready for review and merge as is.

[mgovers]

FINALLY

[Jerry-Jinfeng-Guo]

I will approve the PR once ci is fixed and code cleaned

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
94.8% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud