Test mergin tracking of schema changes to `form_pscis_2023.gpkg`

[NewGraphEnvironment]

mergin may not be able to handle schema changes (ie. extra columns and column type changes) to the form_pscis_2023.gpkg (or any other gpkg) bc geodiff https://github.com/MerginMaps/geodiff can't do it and we should confirm at the start. Shouldn't matter though as long as we don't use form_pscis_2023.gpkg for actual input on the phone in the field and we do the versioning in csvs with git

see https://github.com/NewGraphEnvironment/dff-2022/discussions/154

[NewGraphEnvironment]

seemed to track addition of x and y columns no problem

airvine at Allans-MacBook-Pro in ~/Projects/gis/sern_peace_fwcp_2023
$ mergin status
### Server changes:
### Local changes:

>>> Modified:
M data_field/2023/form_pscis_2023.gpkg
### Local changes summary ###

of note - looks like the form_pscis_2023 layer in the Q project was pointing at a layer inside the gpkg (they always have a name b/c the gpkg is a database) not there (likely due to the layer name it was looking for - form_pscis_v70). Updated with latest form_pscis_2023 (sf defaults to burn the layer as the same name as the gpkg if it is not explicity stated in sf::st_write(layer = whateveryouwanttocallit). Makes me thiink that we could back up original amalgamated form in the gpkg and just not use that layer (can of course have many layers in there...).hmm

[NewGraphEnvironment]

watch for column type changes. pretty sure geodiff can't track those so should confirm what happens...... prob doesn't matter for current use cases but good to know for future