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/package.json

@@ -30,9 +30,13 @@
     "doc-embeddings-query": "tsx scripts/run-doc-embeddings-query.ts",
     "graph-rag": "tsx scripts/run-graph-rag.ts",
     "document-rag": "tsx scripts/run-document-rag.ts",
-    "create-test-pdf": "tsx scripts/create-test-pdf.ts"
+    "create-test-pdf": "tsx scripts/create-test-pdf.ts",
+    "llm:azure-openai": "tsx scripts/run-llm-azure-openai.ts",
+    "llm:openai-compat": "tsx scripts/run-llm-openai-compatible.ts",
+    "llm:mistral": "tsx scripts/run-llm-mistral.ts"
   },
   "devDependencies": {
+    "falkordb": "^5.0.0",
     "nats": "^2.29.0",
     "pdf-lib": "^1.17.1",
     "tsx": "^4.21.0",