CLDR is set up to work with any stock version of the tzdb of a given version or later. While we could make an internal change in a tzdb used for testing with CLDR, it means we would be out of sync with whatever tzdb was being used with it by others. If the tzdb doesn't change, some things will just not work properly, which is why we were hoping that it wouldn't be too much trouble to get a new standard version of the tzdb. It is understandable if there are production constraints that prevent this.