Implementing Powell's Dogleg

[benedict-96]

This pr implements the dogleg method.

We should beforehand clarify a few points:

• should we implement a separate cache for the dogleg? It requires two directional vectors? On wikipedia they are called ${\delta_{gn}}$ and $\delta_{sd}$. At the moment the cache only stores one directional vector (which we call $\Delta x$.
• should linesearch be applied to the final direction?
• how is $s$ chosen in $t\delta_{sd} + s(\delta_{gn} - t\delta_{sd})$?

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.09%. Comparing base (b412a36) to head (39a54f6).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #142      +/-   ##
==========================================
+ Coverage   79.82%   80.09%   +0.27%     
==========================================
  Files          49       49              
  Lines        1234     1236       +2     
==========================================
+ Hits          985      990       +5     
+ Misses        249      246       -3     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.