Sorry for the delay on this.
Some Thoughts on the Evolution of Gothic in TTL
A) Writing out ATL Old Gothic
Whilst some argument can be made for some retention of the Gothic Alphabet in TTL since these Goths are now firmly in the Roman World instead of the periphery I can see them adopting the Roman Script.
Transcribing ATL Old Gothic is fairly similar to OTL modern transcription but with these particulars:
/kw/ would be written [qu]
/hw/ would be [hu]
Since the distinction of /j,w/ is much disputed OTL and not really distinguished the in Vulgar Romances we can be cautious and write them [i,u]
/ŋ/ in clusters would be written [n]
/θ/ may be maintained as
but perhaps becoming [th] when lost. Some case can be made for borrowing [þ] from the AS dialects but any Latin will likely use [th] as they did for the AngloSaxons.
The 5 short vowels are fairly consistent with Roman [a, e, i, o, u] (cf [a, aí, i, aú, u] ).
The long vowels are trickier. Mostly I'd expect them to be doubled and subject to how strong Latin/Romance influences are - eg [ee, oo] vs [ei, ou]; /i:/ could variously be [i, ie, y].
Ironically in this TL it will be sorted out whether 6thC Gothic kept the /ai/ and /au/ dipthongs since the expected digraphs are used for /e, o/; they're often written [ái, áu] OTL where they could be used.
B) Phonological Changes
Whilst your notes indicate that a language called Gothic survives to modern ATL times I get the impression of ebb and flow, and bilingualism, across OTL North Italy before standardisation and demarcation sets in. This I think it worthwhile to consider the influence of Latin (incl Church Latin) and the regional Romance dialects [1]. Perhaps a good analogy could be the development of Old Franconian to Dutch; though the effect of Norman on Late Old English woudl also help.
So I see these phonological changes:
i) palatalisation of [c, g] /k, g/ before /i,j/ and most /e/ to
/tʃ, dʒ/. Thus /ki, gi/ would be written (like Italian
) [chi, ghi].
ii) loss of
/θ/ as a separate phoneme - elised medially and assimilated in clusters
iii) further fricatisation of medial [b, g
] /β, ɣ/ to /v, h/
iv) "schwa-isation" and loss of unstressed vowels
v) loss of medial /h/
C) Grammatical Changes
As with other Germanic languages we will see regularisation of word order, levelling of the verb and noun cases, and creation of articles from determiners. How much depends on the influence of native and biligual Latin/Romance speakers. Do you want retention of the unique Class VII verbs? Dual tenses? etc
[1] Some form of Gallo/North-Italian will probably exist - perhaps as Veneto-Istrian or Illyrian?