mapreduce instead of map + reduce for Jacobians & Hessians

[gdalle]

Another partial answer for #563

DI source

• Replace map + reduce with mapreduce in out-of-place Jacobians and Hessians. As discussed in https://discourse.julialang.org/t/type-instability-of-mapreduce-vs-map-reduce/121136 this makes JET unhappy on 1.10.

DIT source

• Add function_filter argument to test_differentiation and pass it to JET. On 1.10, it is defined as follows to ignore the warning (not needed on 1.11)

function_filter = @nospecialize(f) -> f != Base.mapreduce_empty

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 36.36364% with 56 lines in your changes missing coverage. Please review.

Project coverage is 97.49%. Comparing base (7607ec2) to head (4ca7fa1).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...tionInterfaceTest/src/tests/type_stability_eval.jl	32.53%	56 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #565      +/-   ##
==========================================
- Coverage   98.00%   97.49%   -0.51%     
==========================================
  Files         108      108              
  Lines        4812     4836      +24     
==========================================
- Hits         4716     4715       -1     
- Misses         96      121      +25     

Flag	Coverage Δ
DI	98.66% <100.00%> (-0.01%)	⬇️
DIT	95.20% <34.11%> (-1.49%)	⬇️

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

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