Here in Ether 2:12, Joseph Smith changed hath to has in the printer’s manuscript (as part of his editing of the text for the 1837 edition), but this change was never implemented in the printed edition itself. In the next chapter, however, Joseph’s emendation of hath to has showed up in the 1837 edition:
It may be that here in Ether 2:12 the change to has was simply missed in the typesetting. But one could also argue that it was ultimately decided that hath was the appropriate form when referring to deity (as originally in Ether 2:12) but has when referring to man (as edited in Ether 3:15). One wonders, in other words, if there might be some correlation between the biblical style and references to deity, at least in Joseph Smith’s editing. Nonetheless, there is no consistent evidence for this interpretation of Joseph’s editing since we have examples of all four possibilities:
More generally, there does not appear to be any consistency in Joseph Smith’s editing of hath to has. Nor is the original text itself consistent in its use of hath and has, as we can see in the following contrastive pair (based on the earliest extant text):
So there doesn’t seem to be much reason for the change in Ether 2:12 (which was never implemented), in distinction to the change in Ether 3:15 (which was implemented). For each case of hath or has, the critical text will follow the reading of the earliest textual sources, thus hath in both Ether 2:12 and Ether 3:15.
Summary: Maintain the original hath in Ether 2:12, the reading of the earliest text; in Ether 3:15 the original hath will be restored.