This is an extension of the ID algorithm to graphs containing cycles as described in

Causal Calculus in the Presence of Cycles, Latent Confounders and Selection Bias
Patrick Forré, Joris M. Mooij (2019)
https://arxiv.org/abs/1901.00433v2

Depends on:

• m-separation for ADMGs (this is a special case of d-separation, Implementation to calculate conditional independencies #29/Improve d-separation #135)
• sigma-separation for directed cyclic mixed graphs (Sigma Separation for directed mixed graphs #120)