The big fiboa push - open tasks

[m-mohr]

ŇI feel like for a redeploy we want the following issues and PRs all done:

• All PRs in https://github.com/fiboa/cli/pulls merged or closed, + minor release
• All PRs in https://github.com/vecorel/cli/pulls merged or closed, + v0.3.0 release
• All PRs in https://github.com/vecorel/cli/pulls merged or closed, + v0.3.1 release
• Add older variants to pt #322 (based on PT: add the 2017 to 2022 editions #228)
• Splitting publish-portolan: 38 PRs, what depends on what #269
• Make default FiboaConverter.area_calculate_missing = True #277
• Should we define semantics for id? #282
• Default BaseConverter.use_variant_as_determination = True #284
• 32 rows across the code lists disagree with the HCAT taxonomy #304
• Migrate 7z multifile handling to vecorel-cli #312
• index_as_id repeats across source files, and the uniqueness check cannot see it vecorel/cli#47
• A truncated download is cached and later treated as complete vecorel/cli#46
• Discuss max_dropped_share - really good? vecorel/cli#33
• ie: the 2017-2025 years DAFM still serves, as Anonymous LPIS fieldsoftheworld/harmonized-field-data-catalog#33
• Eight proposed datasets, seven of which are not government records fieldsoftheworld/harmonized-field-data-catalog#32
• ch: one snapshot, no source URL, and cantons missing by construction fieldsoftheworld/harmonized-field-data-catalog#30
• HCAT coverage: what should stay unmapped in de_bb, de_nrw and us_usda_cropland fieldsoftheworld/harmonized-field-data-catalog#22
• Data files are not named after their STAC item in 24 collections fieldsoftheworld/harmonized-field-data-catalog#14
• backfill: pt 2015-2019 (2020-2022 are in PR #3) fieldsoftheworld/harmonized-field-data-catalog#8
• And then I feel we should regenerate and redeploy all datasets ideally in portolan format, ideally with a grouping (official / derived) and those ideally with grouping per country if more than one dataset exists.