Include function suffixes when looking for "Did you mean" suggestions

[nhuurre]

Submission Checklist

• Run unit tests
• Documentation
  • If a user-facing facing change was made, the documentation PR is here:
  • OR, no user-facing changes were made

Since #1024, unknown identifier related errors give a similar identifier as a suggestion but the algorithm only considered typos involving at most three letters. Another likely error is omitting the distribution suffix from a function name:
https://discourse.mc-stan.org/t/how-to-interpret-a-returning-function-was-expected-but-an-undeclared-identifier-was-supplied/34078

Release notes

When encountering an unknown identifier that matches a known suffixed function without its suffix, add the known function name to the error message.

Copyright and Licensing

Copyright holder: Niko Huurre

By submitting this pull request, the copyright holder is agreeing to
license the submitted work under the BSD 3-clause license (https://opensource.org/licenses/BSD-3-Clause)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (35c682b) 89.88% compared to head (ddacb1a) 89.88%.
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1401   +/-   ##
=======================================
  Coverage   89.88%   89.88%           
=======================================
  Files          63       63           
  Lines       10508    10513    +5     
=======================================
+ Hits         9445     9450    +5     
  Misses       1063     1063           

Files	Coverage Δ
src/frontend/Environment.ml	69.49% <100.00%> (+2.82%)	⬆️

[WardBrian]

This is going to end up being based on the order of the lists of suffixes in Utils, right? E.g. if a lpdf is available, it will never suggest lccdf

That’s already an improvement so I’m happy to merge as is, but how would you feel about giving a list of suggestions?

[nhuurre]

If any of them matches then almost all of them are going to match, and I think that's excessive. One option would be to omit _*cdfs and choose either _lpdf/_lpmf or _rng based on the argument types but that adds some complexity.