[Feature] Mean Information Gain (MIG) and Fluctuation Complexity (FC)

[DominiqueMakowski]

@MartynaPlomecka I started taking a look at MIG and I started implemented it (see the change log), don't hesitate to review and if you want after Monday I can show you how to hop-on this PR and push some commits yourself ☺️

Based on https://github.com/MartynaPlomecka/Diversity-and-complexity-analysis-of-EEG-singal

• MIG
• Fluctuation Complexity (FC)
• Multiscale version?
• Use modulable delay ? instead of fixed to 1?

[codecov-commenter]

Codecov Report

Merging #674 (415b655) into dev (f888cf3) will decrease coverage by 0.06%.
The diff coverage is 22.22%.

@@            Coverage Diff             @@
##              dev     #674      +/-   ##
==========================================
- Coverage   53.13%   53.07%   -0.07%     
==========================================
  Files         270      271       +1     
  Lines       12198    12225      +27     
==========================================
+ Hits         6481     6488       +7     
- Misses       5717     5737      +20     

Impacted Files	Coverage Δ
neurokit2/complexity/information_gain.py	19.23% <19.23%> (ø)
neurokit2/complexity/__init__.py	100.00% <100.00%> (ø)
neurokit2/rsp/rsp_simulate.py	98.21% <0.00%> (-0.90%)	⬇️
neurokit2/signal/signal_fixpeaks.py	73.60% <0.00%> (+0.74%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f888cf3...415b655. Read the comment docs.

[DominiqueMakowski]

@MartynaPlomecka I am gonna merge this PR into dev because I want to add these indices in my under-revision structure of chaos study, but please do open a new PR to improve the description etc.

Also, I've exposed the delay which allows to build "words" from non-consecutive samples, potentially allowing to accommodate different sampling rates