Wrap parent of PDMat in Symmetric in eigmin/eigmax

[jishnub]

julia> X = randn(400,600); A = X * X' + I; isposdef(A)
true

julia> S = Symmetric(A);

julia> P = PDMat(S, cholesky(S));

julia> @btime eigmax($P);
  7.598 ms (11 allocations: 1.36 MiB) # master
  4.388 ms (11 allocations: 1.36 MiB) # PR

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (cd28425) 91.48% compared to head (77fff47) 91.48%.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #200   +/-   ##
=======================================
  Coverage   91.48%   91.48%           
=======================================
  Files           9        9           
  Lines         681      681           
=======================================
  Hits          623      623           
  Misses         58       58           

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

[devmotion]

I think it's fine. My only concern is that currently we don't check that a matrix is symmetric when constructing a PDMat, so this could potentially lead to surprising bugs that are hard to discover. But since we already use this pattern in other functions and rely on symmetry, I think it is OK to address this concern in a separate PR.

[devmotion]

Thank you!