Add some more inverse CDF functions

[ColmTalbot]

I've been using statrs for a project I'm working on, and noticed that many distributions don't have an explicit implementation of the inverse cdf when they can be simply implemented.

I wasn't sure what the expectations are for documentation and tests, I added some tests for a couple of them, but thought I should just ask if there is a standard I'm missing before writing a bunch of boilerplate.

If this contribution is welcome/desired, I'm happy to put in any docs/tests.

[YeungOnion]

Thanks for the contribution!

For docs we've been writing the mathematical form for expression for inverse cdf's where possible, it more clearly distinguishes them from search implementations.

I think what you've done in the beta distribution for tests is good.

It's been mentioned before to use a "nice" way to test fwd and inverse cdf since a triple of (parameters, x, p) specifies both ways if not specifying precision, but I'd much prefer having these implemented than having a clearer set of tests.

---

EDIT: specified which distribution the inverse_cdf tests were already written for.

[codecov]

Codecov Report

Attention: Patch coverage is 91.54930% with 12 lines in your changes missing coverage. Please review.

Project coverage is 89.67%. Comparing base (44ccdb9) to head (fb182d3).
Report is 7 commits behind head on master.

Files	Patch %	Lines
src/distribution/chi_squared.rs	0.00%	3 Missing ⚠️
src/distribution/erlang.rs	0.00%	3 Missing ⚠️
src/distribution/beta.rs	96.00%	1 Missing ⚠️
src/distribution/cauchy.rs	96.55%	1 Missing ⚠️
src/distribution/fisher_snedecor.rs	95.45%	1 Missing ⚠️
src/distribution/pareto.rs	93.75%	1 Missing ⚠️
src/distribution/triangular.rs	95.83%	1 Missing ⚠️
src/distribution/weibull.rs	95.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #260      +/-   ##
==========================================
+ Coverage   89.49%   89.67%   +0.18%     
==========================================
  Files          50       50              
  Lines       10851    10992     +141     
==========================================
+ Hits         9711     9857     +146     
+ Misses       1140     1135       -5     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ColmTalbot]

Thanks, I'll put together the documentation, and some tests that verify the forward and backward match to within reasonable precision.

[ColmTalbot]

I wrote some tests and started on docstrings. I'll ping once I finish off the docstrings.

[ColmTalbot]

@YeungOnion this is now ready for review.

[YeungOnion]

This all looks good to me!

Thanks for fixing some of the other docs and catching the precision typo on the test. 10^+14 relative precision isn't any good.