Improve type-stability of sesolve

[albertomercurio]

With the previous PR, I fixed the type instabilities of the sesolve function.

However, a weak type instability was still present at the level of the ODEProblem generation. By weak I mean that the type of the ODEProblem was known up to a Union.

Fixing this will help me on fixing type instabilities for the mesolve and mcsolve functions.

I think that I have introduced a very smat way to deal with multiple dispatch with keyword arguments. For example, the progress_bar argument now supports both a Bool type and a Val type, with the second one as a default to avoid weak type instabilities. In this way, the user can still use progress_bar = true,false, with the price to pay of having weak type instabilities.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 93.50%. Comparing base (8f5444b) to head (34d3740).
Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #191      +/-   ##
==========================================
+ Coverage   93.37%   93.50%   +0.13%     
==========================================
  Files          28       28              
  Lines        2113     2126      +13     
==========================================
+ Hits         1973     1988      +15     
+ Misses        140      138       -2     

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