Refactor likelihood functions

[jamesmbaazam]

This PR closes #48 by ensuring that likelihoods and log-likelihoods are properly calculated. The log argument is removed from offspring_ll() and retained in likelihood() to centralize the (log)likelihood and joint (log)likelihood calculations. The PR also adds tests to check the functionality of the likelihood() function as well as the *_ll() functions to partly address #45.

This PR also fixes a few minor issues, including:

• Clarifying that loglikelihoods, not likelihoods are returned by the *_ll() family of functions.
• Adding new examples and improving some old ones.

[codecov-commenter]

Codecov Report

Merging #58 (6b7364c) into main (8561fc0) will increase coverage by 22.28%.
Report is 8 commits behind head on main.
The diff coverage is 91.37%.

❗ Current head 6b7364c differs from pull request most recent head 9e601a1. Consider uploading reports for the commit 9e601a1 to get more accurate results

@@             Coverage Diff             @@
##             main      #58       +/-   ##
===========================================
+ Coverage   39.95%   62.23%   +22.28%     
===========================================
  Files           8        8               
  Lines         413      429       +16     
===========================================
+ Hits          165      267      +102     
+ Misses        248      162       -86     

Files Changed	Coverage Δ
R/checks.R	53.33% <0.00%> (+26.66%)	⬆️
R/epichains.R	0.00% <ø> (ø)
R/simulate.r	82.46% <66.66%> (+5.84%)	⬆️
R/likelihood.R	100.00% <100.00%> (+100.00%)	⬆️
R/stat_likelihoods.R	100.00% <100.00%> (+42.85%)	⬆️

... and 2 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[sbfnk]

Great stuff - left a few comments.

[sbfnk]

Great

[jamesmbaazam]

Thanks for your thorough review, @sbfnk. Really enjoyed the process. Will merge this after the checks pass.