_biblio_pm_import() returns an inaccurate count of $nids

in the following:

if (!empty($node->nid)) $nids[] = $node->nid;

_biblio_pm_import() has no way of knowing the nid since in the case of a batch API, nodes are actually created after they have been put into the cache, and effectively after _biblio_pm_import() has finished.

I'm sure what the best way to fix this is, but for the sake of returning the right amount of nodes which are due to be created, we could record a zero value nid for unkown nid. At least the count will be correct.