Update read_habitatdata to version 2023

[cecileherr]

Updated code and documentation to version 2023 for:

• read_habitatmap
• read_habitatmap_stdized
• read_habitatmap_terr
• read_habitatstreams
• read_watersurfaces_hab

Added an attribute fix_geom to read_habitatmap (without the st_is_valid control that runs too slow and for which I haven't found a solution yet)

[florisvdh]

Well done @cecileherr, thanks a lot!! ✨

I suggest to revert f9fbf66 (which relates to #179), for the following reasons:

• there are > 350 occurrences of .data$ in the package, so this would only be a start to solve the issue
• the warnings only appear when editing (debugging) {n2khab} functions, not when running them. So currently it's more a developer issue than a user issue. (Still, it would become urgent once {tidyselect} would drop support for .data$, if it ever will).
• most importantly, dropping .data$ provides not enough remediation since tidy selection is not supported by R CMD check; f9fbf66 triggers below note (which is exactly the reason why .data$ was formerly advised in R packages that use tidy selection).

── R CMD check results ─────────────────────────────── n2khab 0.10.1.9000 ────
Duration: 58.6s

❯ checking R code for possible problems ... NOTE
  read_habitatstreams: no visible binding for global variable ‘NAAM’
  read_habitatstreams: no visible binding for global variable ‘BRON’
  read_habitatstreams: no visible binding for global variable
    ‘river_name’
  read_habitatstreams: no visible binding for global variable ‘source_id’
  read_watersurfaces_hab: no visible binding for global variable
    ‘polygon_id’
  read_watersurfaces_hab: no visible binding for global variable
    ‘certain’
  Undefined global functions or variables:
    BRON NAAM certain polygon_id river_name source_id

0 errors ✔ | 0 warnings ✔ | 1 note ✖

Starting with {tidyselect} 1.2.0, tidy selection should work by quoting variable names instead of using .data$; at the same time, we also need to distinguish tidy selection from data masking (see here). We need to have this checked well in various places in {n2khab} when doing this migration, as we have no unit tests in place (yet).

[cecileherr]

I suggest to revert f9fbf66 (which relates to #179), for the following reasons:

commit 39f3976

[florisvdh]

Perfect! 👌