Add scalar benchmark functions

[luward]

We introduce a benchmark set for estimagic consisting of 283 problems for 78 functions. The functions are based on a collection of optimization problem functions by Axel Thevenot. We verified function maths, aswell as the minima using online resources on optimization test problems by the Simon Fraser University and the 2013 paper "A Literature Survey of Benchmark Functions for Global Optimization Problems" by Jamil & Yang . We wrote parameterized unit tests, similarly to those for the "more_wild" problem set, to cover our newly added functions and problems. A detailed description of our work is contained in this repository.

[codecov]

Codecov Report

Merging #271 (10e8bba) into main (fc575ea) will decrease coverage by 0.57%.
The diff coverage is 95.23%.

@@            Coverage Diff             @@
##             main     #271      +/-   ##
==========================================
- Coverage   93.61%   93.04%   -0.58%     
==========================================
  Files         191      193       +2     
  Lines       15357    15705     +348     
==========================================
+ Hits        14376    14612     +236     
- Misses        981     1093     +112     

Impacted Files	Coverage Δ
src/estimagic/benchmarking/more_wild.py	100.00% <ø> (ø)
src/estimagic/benchmarking/scalar_functions.py	100.00% <ø> (ø)
...c/estimagic/benchmarking/get_benchmark_problems.py	94.83% <88.23%> (+0.59%)	⬆️
tests/benchmarking/test_get_benchmark_problems.py	100.00% <100.00%> (ø)
tests/benchmarking/test_scalar_functions.py	100.00% <100.00%> (ø)
tests/optimization/test_tao_optimizers.py	22.68% <0.00%> (-69.08%)	⬇️
src/estimagic/optimization/tao_optimizers.py	25.00% <0.00%> (-45.66%)	⬇️
...ptimization/subsolvers/bounded_newton_quadratic.py	85.08% <0.00%> (-0.88%)	⬇️
...magic/optimization/subsolvers/_trsbox_quadratic.py	84.08% <0.00%> (-0.82%)	⬇️
... and 1 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update fc575ea...10e8bba. Read the comment docs.

[luward]

We introduce a benchmark set for estimagic consisting of 283 problems for 78 functions. The functions are based on a collection of optimization problem functions by Axel Thevenot. We verified function maths, aswell as the minima using online resources on optimization test problems by the Simon Fraser University and the 2013 paper "A Literature Survey of Benchmark Functions for Global Optimization Problems" by Jamil & Yang . We wrote parameterized unit tests, similarly to those for the "more_wild" problem set, to cover our newly added functions and problems. A detailed description of our work is contained in this repository pull request.

[janosg]

I will close this for now because it is incompatible with some of the planned changes in benchmarking (see #495). I kept the files so we can easily re-add this functionality later.