Jiayuan 16f0e49c44 fix(runtimes): persist model-list requests in PostgreSQL to fix multi-instance 404 (#1958) ...

The model-list discovery flow (POST create → poll GET) used an in-memory
map, so in multi-instance deployments the poll could land on a different
pod and return 404. Move the store to a database table so all pods share
the same state.

Changes:
- Add migration 068: model_list_request table with partial index
- Add sqlc queries with FOR UPDATE SKIP LOCKED for atomic PopPending
- Introduce ModelListStorer interface; DBModelListStore for production,
  InMemoryModelListStore for unit tests
- Update model-dropdown and model-picker to show "discovery failed"
  instead of misleading "No models available" on error

Co-authored-by: multica-agent <github@multica.ai>

2026-05-01 09:19:21 +02:00

..

agent

feat(quick-create): gate on daemon CLI version with pre-check + server enforcement (#1857)

2026-04-29 18:44:19 +08:00

db

fix(runtimes): persist model-list requests in PostgreSQL to fix multi-instance 404 (#1958)

2026-05-01 09:19:21 +02:00

protocol

feat(projects): typed project resources + agent runtime injection (#1926)

2026-04-30 14:00:43 +08:00

redact

feat(security): redact sensitive information in agent live output

2026-03-30 23:38:49 +08:00