suggest calling cholesky(Hermitian(A)) for slightly non-Hermitian matrices (#35320)

Author: stevengj

stdlib/LinearAlgebra/src/cholesky.jl

@@ -301,6 +301,9 @@ The triangular Cholesky factor can be obtained from the factorization `F` with:
 The following functions are available for `Cholesky` objects: [`size`](@ref), [`\\`](@ref),
 [`inv`](@ref), [`det`](@ref), [`logdet`](@ref) and [`isposdef`](@ref).
 
+If you have a matrix `A` that is slightly non-Hermitian due to roundoff errors in its construction,
+wrap it in `Hermitian(A)` before passing it to `cholesky` in order to treat it as perfectly Hermitian.
+
 When `check = true`, an error is thrown if the decomposition fails.
 When `check = false`, responsibility for checking the decomposition's
 validity (via [`issuccess`](@ref)) lies with the user.
@@ -354,6 +357,9 @@ The following functions are available for `CholeskyPivoted` objects:
 The argument `tol` determines the tolerance for determining the rank.
 For negative values, the tolerance is the machine precision.
 
+If you have a matrix `A` that is slightly non-Hermitian due to roundoff errors in its construction,
+wrap it in `Hermitian(A)` before passing it to `cholesky` in order to treat it as perfectly Hermitian.
+
 When `check = true`, an error is thrown if the decomposition fails.
 When `check = false`, responsibility for checking the decomposition's
 validity (via [`issuccess`](@ref)) lies with the user.