Add deploymentStart and deploymentEnd to the export CSV in 'Deployments' tab

[martijn-bollen]

Actual: Export CSV exports the following fields: depolymentID, locationID, locationName, latitude and longitude
Expected: Export all fields above and deploymentStart, deploymentEnd

Also allow editing of these new fields upon import.

[Chouffe]

deploymentStart and deploymentEnd are derived values from the media for a given deployment. So it could be useful information for a user to get but not something that one could artificially set. What was your idea about being able to set it?

[wsyxbcl]

@Chouffe Let me add some context here.

There are two slightly different situations:

1. Even if the media timestamps are correct, the first and last media timestamps are not always exactly the same as the actual deployment start and end time. For example, a camera may continue working for some time without object passing by. Users may know the actual setup and collection time from field records, which can be a more accurate deployment range for calculating camera working days.
   (There are also more complicated cases where a camera works discontinuously, though this may be outside the scope of Biowatch. There is a related discussion here: Trap night callendar and deployements tdwg/camtrap-dp#355)

2. The case where media timestamps are not unreliable. For example, the in-camera clock may not have been set correctly, or some metadata may have been lost (and is processed with OCR or a timestamp offset correction), which corresponds to the timestampIssues in Camtrap DP.

So I guess it depends on what BioWatch aims to be. If it is mainly a quick data-wrangling tool, then deriving an initial deployment table from media and letting users continue editing it elsewhere may be enough. But if it aims to be more of a data management system where users manage deployments inside, then preserving and allowing updates to known deploymentStart / deploymentEnd may be useful.

#603 is a relevant issue.