Fix #6.

Only point releases are relevant for dispersal. Observations following areal releases are excluded.

Allow filtering release events from a sit_revents object.

It requires a full sit object, since it needs to extract release events (if following releases is missing) and also extract survey distances from release.

This impacts other functions that uses dispersal, such as flight range, mdt, survival and wild size.

It does not work.

Reverting to how it was before, which worked. I don't know why the slashes are omitted now.

Avoids warning of 'data created with old {dm}'.

Fix #1.

Fix #2.

It essentially removed all egg surveys. Thus, filter observations manually rather than using dm_filter().

Besides, it restricts to point releases.