#149 adjacent gaps broken if not both @extent=unknown

[BODARD Gabriel]

A problem caused by the code that eliminates the second of two adjacent gaps (originally written for DDBDP), which seems to be over-matching. I.e. even if one or both of the gaps has @quantity instead of extent=unknown, the second gap is eliminated. Not good.

[Scott Vanderbilt]

Bump -> Future

[Elli Mylonas]

examples: https://github.com/Brown-University-Library/iip-texts/blob/master/epidoc-files/elus0045.xml
https://github.com/Brown-University-Library/iip-texts/blob/master/epidoc-files/mger0258.xml

[BODARD Gabriel]

@hcayless: do you (a) still plan to implement a fix to this, and (b) hope to do so by September 2020?

[Pietro Maria Liuzzo]

actually, I am not able to reproduce this isse with the examples above and I do not understand why would one encode twice the same gap with extent=unknow. I have tested with

<lb/><gap reason="lost" extent="unknown" unit="character"/><gap reason="lost" extent="unknown" unit="character"/> the second gap should not appear
               <lb/><gap reason="lost" extent="unknown" unit="character"/><gap reason="lost" quantity="4" unit="character"/>  this should be [--- ca.4] 
               <lb/><gap reason="lost" quantity="4" unit="character"/><gap reason="lost" extent="unknown" unit="character"/>   this should be [ca.4 ---]

and in this proposal fix this works like this, making the selector explicit about both the ignored and the checked gap.