Updates diagnostic for computing mixed layer depth

[glwagner]

The diagnostic is based on a buoyancy differential. I used a default 3e-5, which is the recommended potential density difference 0.03 (see for example Treguier et al 2023) divided by a reference density ~1e3.

We should add a test and also limit by the static column depth from Oceananigans, @simone-silvestri maybe you can look this over.

I also added an example which is not literated, but we could consider literating it.

[glwagner]

Hmm looks like this may depend on #284 so let's merge that first

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 47 lines in your changes missing coverage. Please review.

Project coverage is 0.00%. Comparing base (a193aee) to head (211cdfd).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/Diagnostics/mixed_layer_depth.jl	0.00%	47 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #296   +/-   ##
=====================================
  Coverage   0.00%   0.00%           
=====================================
  Files         34      34           
  Lines       1951    1961   +10     
=====================================
- Misses      1951    1961   +10     

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

[navidcy]

I also added an example which is not literated, but we could consider literating it.

That's great. But you also deleted the example that was literated. And then #312 put the non-literated example (that uses GLMakie and sliders and what not) in the docs and made all doc builds fail onwards.

[navidcy]

My only point was that if the literated file is deleted then we should remove it from make.jl otherwise it might trigger mistakes as in #312.