Re: [PATCH 0 of 4] Implement basic sparse op diag and diag op sparse support.

[Jason Riedy]

And I write:
> With the basic sparse & diag (and sparse & permutation) ops, I don't see
> why either diag or sparse should fall back to sparse.

Ok, with one exception.  I have use for the convex combination of
permutation matrices being sparse.  But that I likely should wrap
in a function that also would hide sparse(perm) constructions.

Jason