feat: pluggable image-to-text service with OpenAI vision backend (#1038)

Adds a full-stack image description service: schema, base class,                                                                  
OpenAI backend, gateway dispatch, client APIs (sync/async REST +
websocket), tg-describe-image CLI, IAM capability, and specs.                                                                     
                                                                                                                                    
Closes #879

dev-tools/proc-group/README.md

@@ -111,7 +111,7 @@ Processors that talk to external LLMs or APIs read their credentials
 from env vars, same as in the per-container deployment:
 
 - `OPENAI_TOKEN`, `OPENAI_BASE_URL` — for `text-completion` /
-  `text-completion-rag`
+  `text-completion-rag` / `image-to-text`
 
 Export whatever your particular `group.yaml` needs before running.
 

dev-tools/proc-group/groups/image-to-text.yaml

@@ -0,0 +1,16 @@
+# Image-to-text.  Outbound vision-model calls.  Isolated for the same
+# reason as the LLM group: the upstream API is the most likely thing to
+# need restart (provider changes, model changes, API flakiness).
+
+_defaults: &defaults
+  pubsub_backend: rabbitmq
+  rabbitmq_host: localhost
+  log_level: INFO
+
+processors:
+
+  - class: trustgraph.model.image_to_text.openai.Processor
+    params:
+      <<: *defaults
+      id: image-to-text
+      max_output: 4096