🩹 Fix very slow data/residual plots

[s-weigand]

This reduces plot time of plot_residual (also used inside of plot_overview) from 3min to 3sec.
By default matplotlib tries to find the ideal position to place the legend where it least interferes with the plot itself, but since we add the line plot of the IRF on top of a 2D plot of the data or residual this wastes a lot of time without getting a conclusive result.

Change summary

• 🩹 Hard code legend position in 2D plot so matplotlib does not try to find ideal position

Checklist

• ✔️ Passing the tests (mandatory for all PR's)

[github-actions]

👈 Launch a binder notebook on branch s-weigand/pyglotaran-extras/fix-slow-2d-plot

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (82304e5) 44.94% compared to head (73d0630) 44.94%.

Files	Patch %	Lines
pyglotaran_extras/plotting/plot_residual.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #239   +/-   ##
=======================================
  Coverage   44.94%   44.94%           
=======================================
  Files          27       27           
  Lines        1039     1039           
  Branches      161      161           
=======================================
  Hits          467      467           
  Misses        564      564           
  Partials        8        8           

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

[jsnel]

Who would have thought something seemingly so trivial causes such slowdowns.