Build a bandwidth-efficient incremental sync client against the Open Charge Map API using its compact/camelCase/boundingbox modifiers

Steps

1. Use the boundingbox parameter with paired (lat,lng),(lat2,lng2) corners to scope each sync request to a geographic tile rather than pulling the entire global dataset per call.
2. Pass compact=true to strip expanded reference-data objects (DataProvider, OperatorInfo, StatusType, etc.) from each POI and instead resolve those via IDs against the separate reference-data endpoint, cached once client-side.
3. Pass camelcase=true if the consuming client expects camelCase property names, to avoid a manual field-mapping layer.
4. Use opendata=true when the integration only wants to redistribute OCM's own openly licensed data, since some contributed station data carries additional attribution/usage restrictions.
5. Fetch the reference-data endpoint (operators, connection types, usage types, status types) on a slow cadence (e.g. daily) and join it locally against the compact POI IDs to reconstruct full records without re-fetching reference data on every tile sync.