Ability to work through data gaps or at least visualize unused data

At the moment, gaps in data result in throwing out points both 1 year prior and 1 year after. At PVRW Eivind Sveen (eivind.sveen@ife.no) had discussed a local fork in which he blows through data gaps continuing on to the next year and includes the 2-year slope along with the Rd distribution (just divide by 2 if you have a single year gap, etc).

Alternatively, it would be really nice to highlight in the daily normalized plot which data points (if any) are not taking part in the overall degradation distribution, or are only contributing once (e.g. in the first or last year). For instance, points which are rejected because they don't have a corresponding pair due to aforementioned gap are still shown in the linear plot, and give the impression that they have a 'vote' in the median Rd distribution. Also, first year points do not have a year previous to them, and are only contributing one out of two possible slopes.

Start by tracing the daily normalized plot, linear plot, and degradation distribution handling described in the issue, then compare the referenced local-fork behavior for data gaps. Done means either gaps can be handled across years with the stated slope and distribution behavior, or the plots clearly identify points excluded from or only partly contributing to the distribution.