feat: add agentic-server package — standalone Express LLM service

[pyramation]

Summary

Adds agentic-server — the Express-only equivalent of graphile-llm. This is a standalone Express router that provides agent threads, chat streaming, billing metering, and inference logging without needing PostGraphile or any Graphile dependencies.

Uses @constructive-io/express-context for tenant-scoped database access (the whole reason we built express-context).

Endpoints:

Method	Path	Description
POST	/v1/threads	Create a new conversation thread
POST	/v1/threads/:thread_id/messages	Send messages + get AI response (streaming SSE)
POST	/v1/orgs/:entity_id/threads	Create thread (entity-scoped)
POST	/v1/orgs/:entity_id/threads/:thread_id/messages	Send message (entity-scoped)
POST	/v1/embed	Generate embeddings

Architecture:

Cloud Function → POST /v1/threads/:id/messages → agentic-server
                                                      │
                                  ┌───────────────────┼───────────────────┐
                                  │                   │                   │
                         express-context        OllamaAdapter        Billing
                        (tenant DB context)    (LLM provider)    (quota + usage)

Dependencies (minimal):

• @constructive-io/express-context — tenant context (domain resolution, JWT, pgSettings, withPgClient)
• @agentic-kit/ollama — LLM provider calls (chat + embeddings)
• express (peer dep)
• pg, pg-cache, @pgpmjs/logger — database connectivity and logging

No graphile-anything. The few SQL queries for agent/billing discovery live inside the package with a simple TTL cache (~45 lines).

Review & Testing Checklist for Human

• Verify the router mounts cleanly alongside the existing PostGraphile server (mount createAgenticRouter() at the top of server.ts to confirm no conflicts)
• Test streaming SSE output format matches what the frontend/cloud functions expect (OpenAI-compatible format)
• Verify billing check_quota / record_usage SQL matches what the billing module actually exposes (function signatures)

Notes

• Build passes (pnpm build), 8 unit tests pass (pnpm test)
• Lint is skipped (pre-existing monorepo-wide issue: no eslint flat config for packages/*)
• This extracts the logic from graphql/server/src/middleware/llm-api.ts into a standalone package, but does NOT remove the original llm-api.ts yet — that can happen once this package is verified in production
• The package publishes as bare agentic-server (no scope) per user request

Link to Devin session: https://app.devin.ai/sessions/2b5a29d83d3f478e8d3d972653b4879c
Requested by: @pyramation

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring