Add Hardiness Zones and generic functions for calculating climatologies

[Zeitsperre]

Offshoot issue from #355, related to #1273

The Hardiness Zones indicator (https://en.wikipedia.org/wiki/Hardiness_zone) would be a nice addition to the set of agriclimatic indicators on offer here. The metric helps determine climate-constrained agricultural suitability using a combination of climate factors. USDA is the most common/simple one (https://www.jstor.org/stable/23298457), but there are other versions, including a Canada-specific version developed by Natural Resources Canada (https://doi.org/10.4141/cjps67-044).

The intriguing part of this indicator is that it requires calculations/approaches that could also be applied for calculating SPI/SPEI and other indicators that are currently implemented in https://github.com/monocongo/climate_indices. It might be worthwhile to work on this indicator in part to achieve the objectives of #1273 (or vice versa).

[Zeitsperre]

A recent article/story map using plant hardiness zones: https://ici.radio-canada.ca/info/2023/sirop-erable-rechauffement-climat-niche-production-acericole-cabane-sucre-printemps-seve-quebec/

[aulemahal]

I had the following idea: we could think of "hardiness zones" as a climate classification, instead of an indicator.

Thus, it could be implemented under a new submodule of "climate classifier", which I would define as:

• Inputs are indicators, maybe even statistics of indicators (ex: 30-yr rolling mean)
• The time dimension is not special, classification does not perform temporal aggregation.
• Output is discrete
  • We could maybe have a special way of outputting correct CF attributes for flags.

In addition to hardiness_zones, we could have:

• köppen (yes with the umlaut)
• any others found here : https://en.wikipedia.org/wiki/Climate_classification

[Zeitsperre]

Moving this from milestone. Hardiness indicators are ready, but an agreed-upon approach for calculating climatologies requires a bit more effort.