implement the GET /v1/models/{model_id} endpoint

[git4me8you]

Hi,
trying to run llamabarn with devonthink - worked well in the past but seems DT team implemented more aggressive model checking recently - now tries to check models returned by models endpoint one by one and doesn't get the required response - so models not longer properly selectable in application. DT sends the following request after querying the models endpoint
Regards

GET /v1/models/gemma-4-26b-a4b-q4 HTTP/1.1
Host: 192.168.19.73:2276
Content-Type: application/json
Accept: application/json
User-Agent: DEVONthink/4.3.1 CFNetwork/3860.600.21 Darwin/25.5.0
Accept-Language: de-DE,de;q=0.9
Accept-Encoding: gzip, deflate
Connection: keep-alive

HTTP/1.1 404 Not Found
Connection: close
Content-Type: application/json; charset=utf-8
Server: llama.cpp
Access-Control-Allow-Origin:
Content-Length: 74
{"error":{"message":"File Not Found","type":"not_found_error","code":404}}