fix: correct subregion initialization by name

- Change the method of finding or initializing subregions to use
  the name attribute instead of the id.
- This improves accuracy when synchronizing subregion data, as it
  ensures that subregions are correctly matched by name.
- Eliminates potential issues with duplicate IDs if they happen
  to be reused across different data sources.

lib/tasks/sync_geo_data.rake

@@ -42,8 +42,7 @@ namespace :geo do
     count = 1
 
     subregions.each do |data|
-      subregion = Subregion.find_or_initialize_by(id: data["id"]) do |s|
-        s.name = data["name"]
+      subregion = Subregion.find_or_initialize_by(name: data["name"]) do |s|
         s.region_id = data["region_id"]
       end
       puts "Sync Subregions[#{count}/#{sum}]: [#{subregion.name}]"