Revert "comment out trinary rules"

[mcabbott]

This reverts #60, thus un-reverts #54.

Wants to be run with #61 and JuliaDiff/ForwardDiff.jl#530 before merging.

[codecov-commenter]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.03 🎉

Comparison is base (2001650) 97.86% compared to head (6a7aeef) 97.89%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #62      +/-   ##
==========================================
+ Coverage   97.86%   97.89%   +0.03%     
==========================================
  Files           3        3              
  Lines         187      190       +3     
==========================================
+ Hits          183      186       +3     
  Misses          4        4              

Impacted Files	Coverage Δ
src/rules.jl	100.00% <100.00%> (ø)

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

[devmotion]

I didn't follow the original ForwardDiff issues and registry problems closely - is this guaranteed to not break any existing releases of ForwardDiff?

[mcabbott]

Thanks for looking. I thought so but indeed we should re-check.

ForwardDiff v0.10.18 and below have the error if there is a trinary rule. This is the line upper-bounding their DiffRules in the registry:

https://github.com/JuliaRegistries/General/blob/master/F/ForwardDiff/Compat.toml#L50-L51

The current version is ForwardDiff v0.10.19 which accepts only DiffRules 1.2.1 and up:

https://github.com/JuliaDiff/ForwardDiff.jl/blob/v0.10.19/Project.toml

This is reflected here in the registry:

https://github.com/JuliaRegistries/General/blob/master/F/ForwardDiff/Compat.toml#L22-L23

lines added by the robot JuliaRegistries/General#41858 . So future releases should also work smoothly I believe, not touch old bounds.

So I think that implies this is safe.

[devmotion]

Looks good to me.

[devmotion]

@mcabbott I updated the PR and fixed some tests, do you have any comments? Otherwise I think this PR can be merged.