My response would be that it is included in the Anastasimatarion this way, and that Papa Ephraim (who wrote the orthrography rules article) continues to include that particular thesi in the original form (for instance, in his adaptation of Εν τη Ερυθρά, "In the Red Sea", which I am sure you are familiar with.)
Regarding your first point, just because Ioannis Protopsaltis (d. 1866) wrote this formula with kentemata on page 219 of his Anastasimatarion doesn't necessarily mean that it is orthographically correct to do so. For example, Ioannis Protopsaltis used a petaste in the «Κύριε Εκέκραξα» of First Mode when it is orthographically more correct to use an oligon in that instance. In addition, the purple notes on page 75 of the formula book are an example of a combination that is orthographically incorrect yet frequent in Ioannis's Anastasimatarion. (As an aside, an astute observer will notice that Ioannis Protopsaltis wrote the formula in question with an oligon on page 146 of the same Anastasimatarion.)
Papa Ephraim observed that "the most reliable books for perfect orthography are those with music written by the Three Teachers or their immediate disciples." We therefore examine the orthography employed by Gregorios Protopsaltis (d. 1822), one of the Three Teachers, as well as his immediate disciple Petros Ephesios (d. 1840) in two Doxastika: that of the Archangels («Όπου επισκιάση») and that of St George («Ανέτειλε το έαρ»). In the Syntomon Doxastarion of Petros Peloponnesios as published by Petros Ephesios (Bucharest, 1820), the formula in question is written with an oligon on both page 56 (page 64 of the PDF) and page 185 (page 193 of the PDF). In a manuscript of the Doxastarion of Petros Peleponessios by the hand of Gregorios Protopsaltis that was probably written between 1811 and 1819 (during the period when the New Method was adopted), the formula in question is also written with an oligon on both page 40 and page 127. In another manuscript by Gregorios Protopsaltis, the same formula is presented on page 17 in both the Old Notation and Gregorios' transcription into the New Method, again with an oligon. On the other hand, I was unable to find a single instance of this formula written with kentemata in any nineteenth-century manuscript or in any of the printed editions by Petros Ephesios. The above evidence strongly suggests that writing this formula with an oligon is indeed more correct.