Error by default for pivoted Cholesky of semidefinte matrix

I tried to get the pivoted Cholesky factorization of a positive semidefinite matrix using
```
C = cholesky(diagm([1, 0]), Val(true))
```
However this throws `RankDeficientException(1)`, even though it seems like the matrix was factorized alright?!

It took some time for me to figure out that the following works.
```
C = cholesky(diagm([1, 0]), Val(true). check=false)
```
To me it seems non-intuitive having to turn off the check for something that should work.

According to the docs one should also be able check `issuccess(C)`, however this throws
`ERROR: MethodError: no method matching issuccess(::CholeskyPivoted{Float64,Array{Float64,2}})`

Here are the docs:

```
cholesky(A, Val(true); tol = 0.0, check = true) -> CholeskyPivoted

Compute the pivoted Cholesky factorization of a dense symmetric positive
semi-definite matrix A and return a CholeskyPivoted factorization. The
matrix A can either be a Symmetric or Hermitian StridedMatrix or a perfectly
symmetric or Hermitian StridedMatrix. The triangular Cholesky factor can be
obtained from the factorization F with: F.L and F.U. The following functions
are available for CholeskyPivoted objects: size, \, inv, det, and rank. The
argument tol determines the tolerance for determining the rank. For negative
values, the tolerance is the machine precision.

When check = true, an error is thrown if the decomposition fails. When check
= false, responsibility for checking the decomposition's validity (via
issuccess) lies with the user.
```

Another inconsistency is that the docs for `LAPACK.pstrf!` says that it works for positive-definite matrices while the [LAPACK docs](https://software.intel.com/en-us/mkl-developer-reference-c-pstrf) says that it works for positive semidefinite matrices. If it's not a typo, a remark in the docs could be in order.