MNLE with experimental conditions

[janfb]

changes:

• rename tutorial 14
• add new tutorial 17 for iid data in decision-making context: MNLE, and inference with experimental conditions (by conditioning the on the parameters representing the experimental conditions and running MCMC)
• test for MNLE with experimental conditions
• add separate train method for MNLE to have the correct return type

[codecov]

Codecov Report

Merging #829 (68b3a02) into main (86e73a9) will increase coverage by 0.98%.
Report is 21 commits behind head on main.
The diff coverage is 85.71%.

@@            Coverage Diff             @@
##             main     #829      +/-   ##
==========================================
+ Coverage   74.59%   75.58%   +0.98%     
==========================================
  Files          80       80              
  Lines        6236     6254      +18     
==========================================
+ Hits         4652     4727      +75     
+ Misses       1584     1527      -57     

Flag	Coverage Δ
unittests	75.58% <85.71%> (+0.98%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files Changed	Coverage Δ
sbi/inference/snpe/snpe_c.py	75.36% <0.00%> (ø)
sbi/inference/posteriors/mcmc_posterior.py	85.39% <100.00%> (+0.56%)	⬆️
sbi/inference/snle/mnle.py	85.00% <100.00%> (+1.66%)	⬆️

... and 7 files with indirect coverage changes

[michaeldeistler]

I did not read through the new tutorials, but feel free to merge and I'll try to have a look asap.

Thanks!