Add support for High-Value Datasets and dynamic data

[jakubklimek]

We need to add support for HVDs into forms.

• [x] Change JSON-LD context of produced JSON-LD files to https://ofn.gov.cz/dcat-ap-cz-rozhraní-katalogů-otevřených-dat/2024-05-28/kontexty/rozhraní-katalogů-otevřených-dat.jsonld according to the new version of the specification
• [x] Add Solr Core for HVD categories (loading of the contents ready in pipelines) so that they can be searched and autocompleted in the form. Not all terms have both English and Czech labels, some have only English ones - this will be fixed in September, we can copy the English ones to the Czech fields in the meantime, if there is no fallback in the forms themselves
• [x] Add Multi-select for Dataset type, similar to Themes, containing "High-Value dataset", "Dynamic data", "Data from public registries", corresponding to právní_předpis with values http://data.europa.eu/eli/reg_impl/2023/138/oj, https://www.e-sbirka.cz/eli/cz/sb/1999/106/2024-01-01/dokument/norma/cast_1/par_3a/odst_6 and https://www.e-sbirka.cz/eli/cz/sb/1999/106/2024-01-01/dokument/norma/cast_1/par_5a/odst_1 respectively.
• [x] Add Multi-select for Distribution type, similar to Themes, containing "High-Value dataset", "Dynamic data", "Data from public registries", corresponding to právní_předpis with values http://data.europa.eu/eli/reg_impl/2023/138/oj, https://www.e-sbirka.cz/eli/cz/sb/1999/106/2024-01-01/dokument/norma/cast_1/par_3a/odst_6 and https://www.e-sbirka.cz/eli/cz/sb/1999/106/2024-01-01/dokument/norma/cast_1/par_5a/odst_1 respectively. In case of a data service, these values will also be output for the data service.
• [x] Add Multi-select High Value Dataset Category to Dataset, similar to Themes, with search and autocomplete connected to the new solr core, filling the kategorie_hvd property of a dataset. This value will be also output for Data Services, if selected as distribution type

Add following validations:

• [x] If HVD is selected in Dataset, HVD Category needs to be selected, and at least one of the distributions needs to be marked as HVD as well.

Names and descriptions can be taken from the current dataset list and detail.