new class of PL vectorizations

[corybrunson]

As in Peter Bubenik's lab worksheets, discrete PLs with common and constant support can be encoded as (sparse) matrices, especially for linear-algebraic uses. Since the functions pl_vectorize() and pl_devectorize() already rely on vector/matrix attributes, it would be a small leap to have them convert between class 'Rcpp_PersistenceLandscape' and a new class of (sparse) matrices of vectorized landscapes.

If this class is used, then either (a) most other pl_*() will need to be adapted to handle this class, (b) a parallel set of pv_*() functions will need to be written, or (c) the low-level pl_*() functions will need to be rewritten to use method dispatch.

[corybrunson]

Discussion with colleagues leads me to think that these vectorizations (and matrices of them) can be left unclassed. But there perhaps should be an option to store them as sparse matrices.