feat: add Docker entrypoints, LLM providers, pipeline hardening, workbench pages

Phase 9 — four parallel workstreams:

- Stream A: 14 Docker entrypoints for containerized deployment
- Stream B: Pipeline hardening — robust JSON parsing, LLM retry logic,
  consumer negative-ack, FalkorDB test import fix
- Stream C: Azure OpenAI, OpenAI-compatible, and Mistral LLM providers
- Stream D: Workbench Prompts, Token Cost, Knowledge Cores pages +
  Settings feature switches

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

ts/scripts/run-llm-azure-openai.ts

@@ -0,0 +1,18 @@
+/**
+ * Start the Azure OpenAI text-completion service.
+ *
+ * Usage: AZURE_TOKEN=... AZURE_ENDPOINT=... pnpm tsx scripts/run-llm-azure-openai.ts
+ *
+ * Env:
+ *   NATS_URL          (default: nats://localhost:4222)
+ *   AZURE_TOKEN       (required)
+ *   AZURE_ENDPOINT    (required)
+ *   AZURE_MODEL       (default: gpt-4o)
+ *   AZURE_API_VERSION (default: 2024-12-01-preview)
+ */
+import { run } from "../packages/flow/src/model/text-completion/azure-openai.js";
+
+run().catch((err) => {
+  console.error("Azure OpenAI LLM service failed:", err);
+  process.exit(1);
+});