I am certain that the majority of iFly users today are tablet-based customers, not iFly-device customers. As you yourself point out, it's far easier for users to come to iFly either via a tablet device they already own or an inexpensive tablet they buy for iFly use but that also gives them other uses outside of iFly, rather than a more-expensive device that can only be used for iFly. From a business perspective, it seems to make sense to satisfy that larger user base first.
From a development perspective, the tablet platforms provide the fastest prototyping, as well. Android and Windows 10 builds can be turned around in lminutes/hours and put out to the beta communitry for immediate testing on a wide variety of devices. When the focus is on trying to get new code working and stable, this is an effective and efficient approach to finding and stamping out bugs. (Apple has a review cycle each time new code is released, so there is extra time involved turning around new versions, there.) Optimizing the code often comes later in the process, and sometimes that is the point where it becomes apparent that a new feature will or will not be possible on older hardware.