Use LogDensityProblemsAD with ADTypes

[sethaxen]

LogDensityProblemsAD has ADTypes integration for defining gradients and hessians of the log-density. DynamicPPL.jl then customizes this integration for DynamicPPL.LogDensityFunction (see e.g. https://github.com/TuringLang/DynamicPPL.jl/blob/master/ext/DynamicPPLForwardDiffExt.jl).

Switching to use LogDensityProblemsAD gradient and hessian functions instead of Optimization's ensures that when such customizations are made to improve performance or work around some issue unique to that LogDensityProblem, we then do the right thing. A potential downside is that Optimization.jl may support more ADs than LogDensityProblemsAD.jl, and we would not support these extra ADs. However, a simple workaround is that a user could pass Base.Fix1(LogDensityProblems.log_density, prob) to pathfinder or multipathfinder, which would not be identifiable as a LogDensityProblem and would then result in Optimization's ADTypes integration being used.

Edit: relates #93

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.83%. Comparing base (3b29881) to head (b9ee54a).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #198      +/-   ##
==========================================
+ Coverage   80.61%   80.83%   +0.21%     
==========================================
  Files          13       13              
  Lines         614      621       +7     
==========================================
+ Hits          495      502       +7     
  Misses        119      119              

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