Add coefplot and groupedcoefplot for StatisticalModel

[getzze]

The new dependency is minimal (StatsModels), test dependencies are a bit bigger (CategoricalArrays, DataFrames, GLM).
Tested with the gr, pyplot and plotly backends.
It was tested and working with GLM.jl, MixedModels.jl and RobustModels.jl models. I don't know other packages that use the StatsModels API.

[getzze]

coefplot (cp1 from tests in runtests.jl)

groupedcoefplot (slightly different cp5 from tests in runtests.jl)

[sethaxen]

Thanks for the PR! I'm happy to have this functionality in this package. It's a lot of code though, so we'll probably need a few rounds of review.

A few general comments first:

• I know StatsPlots doesn't consistently do this (yet), but it would make the code more readable if the function calls/types were prefixed with the package name where they're defined (or implemented, if we rely on the implementation)
• Can you make your utility function names more descriptive? This will make it easier for me to follow the code.
• You've essentially implemented a forest plot. This is generally useful apart from plotting coefficients of statistical models. Would it make sense to refactor to a forestplot primitive and a coefplot that extracts the coefficients and passes them to forestplot?

[getzze]

Hi, sorry for the long delay. I had some time so I refactored the code, let me know if it is clearer.
The util functions to parse the formula terms are very specific so I didn't comment them fully.

[codecov-commenter]

Codecov Report

Patch coverage: 71.84% and project coverage change: +11.24 🎉

Comparison is base (f14f0b0) 26.52% compared to head (115ef3c) 37.77%.

Additional details and impacted files

@@             Coverage Diff             @@
##           master     #491       +/-   ##
===========================================
+ Coverage   26.52%   37.77%   +11.24%     
===========================================
  Files          20       21        +1     
  Lines        1248     1660      +412     
===========================================
+ Hits          331      627      +296     
- Misses        917     1033      +116     

Impacted Files	Coverage Δ
src/StatsPlots.jl	33.33% <ø> (ø)
src/statsmodels.jl	71.84% <71.84%> (ø)

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.