Reframing as return maximization

[Foggalong]

Reframing the problem as return maximization, rather than risk minimization. Initial reframing done, with the note that this gives us a slightly more complicated set of $\lambda$ to work through. Tasks to do:

• Rewrite previous notebook to explain the difference in frames
• Check whether there's still numerical error present at $\kappa = 0$
• Better explain the idea of $z$ pushing down for the $\sqrt{w^{T}\Omega w}$ term
• Correct footnotes (objectives + order have now changed)