Nice to see the Latins listed separately. Not so nice to see the uncaught mistakes. E.g.:
The reading of the (Arian) Gothic manuscript is completely obliterated, saei (ὅς, "who") is an editorial decision in the printed text.
Origen's text is quia ("that"), a subordinator in the sentence not qui (ὅς, "who").
Cyril certainly reads θεός in multiple places, whereas the scholia in Oecumenius lacks sufficient context to judge the reading.
The Harkleian Syriac is listed incorrectly (it is either θεός or ὃς θεός).
While the Armenian is listed, there is some debate over the translation of Zohrab's 1805 critical edition, as it can technically (if not reasonably) be interpreted to support the reading θεός due to the construction of the clause preceding the relative.
The apparatuses utilize printed texts for many of the versions including the Armenian. The 11th/12th century Greek-Armenian Diglot 256, for instance, does contain ա͞ծ, the contraction of աստուած, "God" (Եւ յայտնի իսկ մե́ծ է խորհուրդն ա͞ծպաշտութեան. ա͞ծ որ երեւեցա́ւ): great is the sacrament of piety: God, who was manifest in the flesh.
F and G still retain the mark of contraction through the traverse is missing (i.e., error for θεός), which is a line not found anywhere else over ὅς.
Uncials 056, 075, 0142, 0150 and 0151 are omitted from the evidence for θεός.
The Georgian and Slavonic are also omitted from the evidence for θεός.
Also not sure Euthalius quotes the passage, though a strong case could be made from the manuscripts.
Not to mention the trouble with the testimony of codices A and C. Should at least be marked "vid," as problems are present in both texts.
I've never seen l599 or Eutherius listed anywhere else for ὅς. I tried to verify l599 but unfortunately the images are not currently available. It has been removed in the UBS4 and replaced with l60 for ὅς and "l597 ὃς θεός."