Move ad.jl to DynamicPPL

[sunxd3]

Twin PR from DynamicPPL.jl
Address: #2141

Description:

• getADBackends are renamed to getADType and moved to Inference module as well as LogDensityProblemsAD.ADgradient(ℓ::LogDensityFunction)
• The LogDensityProblemsAD.ADgradient(adtype, ℓ) specific to RD and FD are moved to DynamicPPL
• The idea is that with DynamicPPL, call to ADgradient must also gives the adtype, in Turing, we just use the adtype from the algorithm

Will finish up with the test, once DynamicPPL PR is merged

[sunxd3]

looks like TuringLang/AdvancedMH.jl#88 is breaking things

[coveralls]

Pull Request Test Coverage Report for Build 7935140085

Details

• -9 of 9 (0.0%) changed or added relevant lines in 1 file are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage remained the same at 0.0%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/mcmc/Inference.jl	0	9	0.0%

Files with Coverage Reduction	New Missed Lines	%
src/mcmc/Inference.jl	1	0.0%

Totals
Change from base Build 7935132657:	0.0%
Covered Lines:	0
Relevant Lines:	1367

---

💛 - Coveralls

[codecov]

Codecov Report

Attention: 9 lines in your changes are missing coverage. Please review.

Comparison is base (cf647b1) 0.00% compared to head (a303b9b) 0.00%.

❗ Current head a303b9b differs from pull request most recent head 7ee7314. Consider uploading reports for the commit 7ee7314 to get more accurate results

Files	Patch %	Lines
src/mcmc/Inference.jl	0.00%	9 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff           @@
##           master   #2158   +/-   ##
======================================
  Coverage    0.00%   0.00%           
======================================
  Files          22      21    -1     
  Lines        1385    1367   -18     
======================================
+ Misses       1385    1367   -18     

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

[sunxd3]

This seems to causing method redefine errors, ref #2173

[yebai]

Looks good to me, @sunxd3! Can you add a note about this change to History.md and release info, so some users depending on Turing internals do not get surprised.