feat: LLM entity + fact extraction pipeline (Zep paper alignment) (#48)

## Changes

### Entity Extraction
- Switch from WikiLinkFallbackExtractor to LlmEntityExtractor when LLM_ENDPOINT set
- `clean_llm_response()`: strips `<think>` tags, markdown fences, extracts JSON
- Handle array responses (Ollama returns `[...]` not `{entities: [...]}`)
- EntityType custom Deserialize: unknown variants → Unknown (no crash)
- Increase timeout 30s→90s, max_tokens 500→1500 for reasoning models
- Graceful reflection fallback: keep entities if verification fails

### Fact Extraction (NEW)
- LlmFactExtractor: LLM-based relationship extraction between entity pairs
- Validates source/target against known entity list (drops hallucinated edges)
- Same robust JSON cleaning for reasoning models + Ollama
- IngestWorker auto-selects LLM vs Simple based on LLM_ENDPOINT env

### K8s Deployment
- Add `command: ["/app/mem"]` (fix args replacing CMD)
- Add LLM_ENDPOINT, LLM_MODEL env vars for in-cluster LLM

## E2E Tested (local Ollama qwen2.5:3b)
- 12 entities extracted (person, tool, concept, organization)
- 5 edges with relationships and facts
- 781 tests pass

## Zep Paper Alignment (§2.2)
- Entity extraction + resolution (§2.2.1)
- Fact extraction between entity pairs (§2.2.2)
- Temporal edge invalidation ready (t_valid/t_invalid schema)
- Reflection verification (§2.2.1, graceful fallback)

---------

Co-authored-by: rock <[email protected]>
Reviewed-on: #48
Co-authored-by: poimen <[email protected]>

k8s/infra/databases/memory-db.yaml

@@ -1,8 +1,6 @@
----
-# CNPG Postgres cluster for Poimen Memory system (GitOps, declarative extensions).
-# 2 instances, pgvector 0.7.0 via spec.extensions (not manual CREATE EXTENSION).
-# Storage: 10Gi longhorn, consistent with temporal-db.yaml.
-# No manual psql needed — all via git/ArgoCD.
+# Dedicated CNPG Postgres for Poimen Memory (GitOps, wave 2).
+# Matches homelab/k8s/infra/databases/memory-db.yaml — single source of truth.
+# CNPG generates secret `memory-db-app` + service `memory-db-rw` in ns poimen.
 apiVersion: postgresql.cnpg.io/v1
 kind: Cluster
 metadata:
@@ -13,20 +11,6 @@ metadata:
 spec:
   instances: 2
   imageName: ghcr.io/cloudnative-pg/postgresql:16.2
-  enableSuperuserAccess: false
-  storage:
-    size: 10Gi
-    storageClass: longhorn
-  resources:
-    requests: { memory: "512Mi", cpu: "250m" }
-    limits: { memory: "2Gi", cpu: "1" }
-  affinity:
-    podAntiAffinityType: preferred
-    topologyKey: kubernetes.io/hostname
-    tolerations:
-      - key: node-role.kubernetes.io/control-plane
-        operator: Exists
-        effect: NoSchedule
   bootstrap:
     initdb:
       database: memory
@@ -34,25 +18,19 @@ spec:
       encoding: UTF8
       localeCollate: C
       localeCType: C
-  monitoring:
-    enabled: true
-    podMonitorTemplate:
-      spec:
-        interval: 30s
-        scrapeTimeout: 10s
----
-# Database resource with pgvector extension (declarative, git-managed).
-# CNPG 1.30.0+ supports this via spec.extensions on the Database CRD.
-# Ensures pgvector is installed and available for HNSW indexing.
-apiVersion: postgresql.cnpg.io/v1
-kind: Database
-metadata:
-  name: memory
-  namespace: poimen
-spec:
-  cluster:
-    name: memory-db
-  owner: app
-  extensions:
-    - name: vector
-      ensure: present
+      postInitApplicationSQL:
+        - "CREATE EXTENSION vector;"
+  enableSuperuserAccess: false
+  resources:
+    requests: { memory: "512Mi", cpu: "250m" }
+    limits: { memory: "2Gi", cpu: "1" }
+  storage:
+    size: 20Gi
+    storageClass: longhorn
+  affinity:
+    podAntiAffinityType: preferred
+    topologyKey: kubernetes.io/hostname
+    tolerations:
+      - key: node-role.kubernetes.io/control-plane
+        operator: Exists
+        effect: NoSchedule