Feature Request: Integrate fGES and XGES into causal-learn

[Jessez2]

Hello maintainers,
Thank you so much for developing and maintaining causal-learn, a great unified package for causal discovery.
I have a feature request that I believe will strongly benefit users:
Please integrate two important improved GES-style algorithms into causal-learn:

1. Fast Greedy Equivalence Search (fGES)
2. Extremely Greedy Equivalence Search (XGES)
   These methods are widely used in modern causal discovery and can greatly improve both speed and accuracy over the standard GES, especially on high-dimensional and dense graphs.
   Relevant references
   XGES:
   Achille Nazaret and David Blei. Extremely Greedy Equivalence Search. UAI 2024.
   https://openreview.net/forum?id=2gIMXx9UxRN
   fGES:
   Joseph Ramsey, Madelyn Glymour, Ruben Sanchez-Romero, and Clark Glymour. A million variables and more: the fast greedy equivalence search algorithm for learning high-dimensional graphical causal models. International Journal of Data Science and Analytics, 3(2):121–129, 2017.

Thank you very much for your consideration!

[kunwuz]

Thanks! This is very helpful. We will try to incorporate different variants of GES. For FGES, perhaps you could use Tetrad for now, since it is currently available there.