Make DynamicHMC and Turing support into extensions

[sethaxen]

On Julia v1.9 and newer versions, this PR makes the DynamicHMC and Turing support into extensions. This will make it easier to keep compats up-to-date and allows for precompilation of the glue code.

On older Julia versions, Requires is used to conditionally load the extensions.

Until JuliaRegistries/Registrator.jl#397 is merged, this will need to be registered using LocalRegistry#master. Since this usage isn't documented, it's probably best to wait for the above PR to be merged so we can use JuliaRegistrator.

[codecov-commenter]

Codecov Report

Base: 92.64% // Head: 92.65% // Increases project coverage by +0.01% 🎉

Coverage data is based on head (71a788c) compared to base (b5d1917).
Patch coverage: 100.00% of modified lines in pull request are covered.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #114      +/-   ##
==========================================
+ Coverage   92.64%   92.65%   +0.01%     
==========================================
  Files          13       13              
  Lines         571      572       +1     
==========================================
+ Hits          529      530       +1     
  Misses         42       42              

Impacted Files	Coverage Δ
ext/DynamicHMCExt.jl	100.00% <100.00%> (ø)
ext/TuringExt.jl	96.15% <100.00%> (ø)
src/Pathfinder.jl	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

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

[sethaxen]

The following cell in the docs build fails:

Pathfinder.jl/docs/src/examples/turing.md

Lines 105 to 107 in 7d2ce56

	samples_transformed = reduce(vcat, fun.transform.(samples)')
	varnames = Pathfinder.flattened_varnames_list(model)
	chns = MCMCChains.Chains(samples_transformed, varnames)

This happens because flattened_varnames_list is defined in the extension code and is not accessible to the user. We should check if this functionality had made it to Turing or DynamicPPL yet so that we can remove it.