Scaled and Shifted

[cscherrer]

Generally the idea is

• Have a Shiftable trait
• For some distributions like Gaussian , some_dist.shifted(dx) gets absorbed and just gives another instance of the distribution
• When that's not an option, we wrap with Shifted
• There's also a ShiftedSuffStat that shifts the data before feeding into the suffstat of the parent

It will be similar for Scaled

[codecov]

Codecov Report

Attention: Patch coverage is 49.11111% with 229 lines in your changes missing coverage. Please review.

Project coverage is 80.82%. Comparing base (f6b3d9b) to head (f5aa147).
Report is 1 commits behind head on release/v0.19.0.

Files with missing lines	Patch %	Lines
src/data/stat/scaled.rs	0.00%	32 Missing ⚠️
src/dist/shifted.rs	59.72%	29 Missing ⚠️
src/dist/scaled.rs	69.66%	27 Missing ⚠️
src/data/stat/shifted.rs	0.00%	25 Missing ⚠️
src/traits.rs	4.00%	24 Missing ⚠️
src/dist/gaussian.rs	48.00%	12 Missing and 1 partial ⚠️
src/dist/cauchy.rs	50.00%	12 Missing ⚠️
src/dist/gev.rs	57.14%	12 Missing ⚠️
src/dist/laplace.rs	50.00%	12 Missing ⚠️
src/dist/uniform.rs	50.00%	9 Missing ⚠️
... and 7 more

Additional details and impacted files

@@                 Coverage Diff                 @@
##           release/v0.19.0      #16      +/-   ##
===================================================
- Coverage            81.49%   80.82%   -0.68%     
===================================================
  Files                   93       97       +4     
  Lines                18810    19241     +431     
  Branches             18810    19241     +431     
===================================================
+ Hits                 15329    15551     +222     
- Misses                3382     3591     +209     
  Partials                99       99              

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[cscherrer]

Some things to confirm, if you could:

1. Every continuous univariate distribution should implement both Scalable and Shiftable
2. Distributions that implement either of these manually should have the corresponding test macros included

I checked these but it's manual so a second look would help 🙏

[schmidmt]

I have a few questions, but overall, this looks great! I love all the tests!