backend: AI chat agent with Veil PHI safeguard (providers, /chat, import)

Real LLM chat replacing the mock, backend-centric per the plan:

- Multi-provider API-key mode (OpenAI / Anthropic / Gemini via the AI SDK) plus
  local Ollama (OpenAI-compatible endpoint). Provider is derived from the
  picked model id; the matching stored key is used. New user_ai_settings table
  holds per-user config with provider API keys encrypted at rest (AES-256-GCM,
  src/lib/crypto.ts, keyed by AI_CREDENTIALS_KEY).
- POST /api/ai/config (get/put, secrets never returned), POST /api/ai/test
  (Ollama ping / key presence), POST /api/ai/import (approved migration commit,
  re-validated server-side, reuses the audited patient service).
- POST /api/chat: streamText agent with tools (getPatient, getPatientLabs,
  searchPatients, previewImport). Real record data streams to the clinician as
  custom data parts (cards) while the model sees only Veil-redacted results.
- Veil (src/services/ai/veil.ts): de-identifies patient identifiers to tokens
  before external calls, resolves tokens on tool args, and rehydrates the final
  answer. Bypassed for local Ollama. External mode runs non-streamed so the
  rehydrated text is correct. Every call is audited (provider + Veil level).
- Shared role-scoping helpers extracted to src/lib/role-scope.ts (reused by the
  patient routes and chat tools so visibility rules match).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

backend/package.json

@@ -20,6 +20,11 @@
     "db:push": "drizzle-kit push"
   },
   "dependencies": {
+    "@ai-sdk/anthropic": "^3.0.84",
+    "@ai-sdk/google": "^3.0.82",
+    "@ai-sdk/openai": "^3.0.71",
+    "@ai-sdk/openai-compatible": "^2.0.50",
+    "ai": "^6.0.204",
     "better-auth": "^1.6.13",
     "cors": "^2.8.6",
     "dotenv": "^17.4.2",