- define NLMOD_CACHE_OPTIONS
- set defaults (identical to current caching defaults)
- add context manager for cache options
- add functions to set, get and reset options.

- add check for setting NLMOD_CACHING_OPTIONS
- add logging when global config modifies default
- do not check hashes if global option overrides cache decorator setting
- add explicit dataset coordinate check, useful if hash checks are disabled.

…s None.

- use assert_identical
- invalidate cache when explicit check fails

Add caching options

- allows for using different option for parsing paleochannels without having to run all methods separately.

- e.g. `obs_type=["stage", "lak"]`

Add sources for "Hunze en Aa's"
Change source of level-areas for "Noorderzijlvest"
Change source of watercourses of "Aa en Maas", "De Dommel" and "Schieland en de Krimpenerwaard"
Change index-name of watercourses for "Rivierenland" (from "code" to "CODE", the default)

And upload summer_stage.png and winter_stage.png for documentation

- accessing gwf.modelgrid arrays is insanely slow for some reason.
- add a different strt computation when strt=None. Default is now to use model surface level instead of 0.0.

Pass geotop layers to get_combined_layer_models

allow getting layer top/botm from ds

allow passing multiple lake observation types as a list of str

* Add rename_layers_to_old_names

* Name argument rename_layers_to_version_2_2_2  instead of rename_layers_to_old_names

* move files to regis directory and add to pyproject

* add regis to pyproject

* excel -> csv and add mv

* codacy

---------

Co-authored-by: Onno Ebbens <onnoebbens@gmail.com>

* add functions to check ds or da grid type
- is_structured
- is_vertex
- is_layered

* add _get_empty_layered_da
- add method to get empty data array with new layer coord but maintaining other dims/coords from source data array

* use new method to get empty da when combining layers

* fix combine_layers to work for vertex datasets

* improve combine_layers
- only drop datavars with layer coord
- add the other datavars back to final combined result
- lazy formatting in logging

* add import

* ruff

* ruff

* Add a GridTypeDims enum
- specificies dimensions for grid types

* ruff

* add test for combining layers unstructured

* add option to return reindexer:
- no longer add reindexer to attributes (this was annoying when saving the dataset)

* make enum even more complicated
- add parse_dims class method
- use this in is_structured and is_vertex grid-methods

* codacy

* docstring