Consecutively supplied words and supplied in abbreviated words

[zeichman]

Presently, consecutive words that are supplied are given their own individual brackets:
There are three display issues that should be fixed, allof which are in evidence in https://library.brown.edu/iip/viewinscr/mtgr0001/

First, and more simple, is evidenced in line 5: [et] [P] [Sep-]
In this case, we have three consecutive words that are all supplied, but each word is separated by its own square bracket. This should instead display as: [et P Sep-]. This is because each of these is separated by its own attribute, with a space in between.

Second, and more complicated, is that abbreviated words with part of it supplied are not always having the excerpted portion bracket correctly. Consider line one of the same inscription. It presently displays as: Imp(eratori) Caes [Lucio] [Septimio]. The issue here is that Caes should display as Ca[es(ari], but it cannot because cannot be nested within .

Third, and cumulative, is that when abbreviated words with partial supplieds are adjacent to fully supplied words, they presently display as follows (again, line one of the same inscription): Imp(eratori) Caes [Lucio] [Septimio]
It should instead display as Imp(eratori) Ca[es(ari) Lucio Septimio] - in this case a single bracket links everything from the "e" of Caes(ari) to the "o" of Septimio.

[zeichman]

Related to this is the issue of how unrestored lacunae (<gap .../> next to supplied letters gets displayed. For instance, haif0020 displays κα[ὶ][---] instead of κα[ὶ ---].