feat(deploy): wire qdrant_api_key end-to-end, add CLI storage flags, docker services

- VektoriConfig gains qdrant_api_key field; factory.py passes it to QdrantBackend
- Vektori.__init__ exposes qdrant_api_key so callers don't need to build VektoriConfig
- CLI: --storage-backend / --database-url / --qdrant-api-key added to all commands;
  `vektori config` can now save storage settings to ~/.vektori/config.json
- docker-compose.yml adds Neo4j 5 and Qdrant services alongside Postgres
- README: remove "No Neo4j, no Qdrant" claim; document all four backends + pip extras

Author: laxmanclo

README.md

@@ -25,7 +25,7 @@ EPISODE LAYER (L1)   <- patterns auto-discovered via graph traversal.
 SENTENCE LAYER (L2)  <- raw conversation. Sequential NEXT edges. The full story.
 ```
 
-Search hits Facts, graph discovers Episodes, traces back to source Sentences. One database, Postgres or SQLite. No Neo4j, no Qdrant, no infra drama.
+Search hits Facts, graph discovers Episodes, traces back to source Sentences. SQLite by default — swap to Postgres, Neo4j, or Qdrant when you're ready to scale.
 
 ---
 
@@ -42,7 +42,10 @@ Still improving. Run your own in [`/benchmarks`](benchmarks/).
 ## Install
 
 ```bash
-pip install vektori
+pip install vektori                      # SQLite + Postgres
+pip install 'vektori[neo4j]'             # + Neo4j support
+pip install 'vektori[qdrant]'            # + Qdrant support
+pip install 'vektori[neo4j,qdrant]'      # all backends
 ```
 
 No Docker, no external services. SQLite by default.
@@ -192,16 +195,53 @@ v = Vektori()
 # PostgreSQL + pgvector — production scale
 v = Vektori(database_url="postgresql://localhost:5432/vektori")
 
+# Neo4j — native graph traversal for Episode layer
+v = Vektori(
+    storage_backend="neo4j",
+    database_url="bolt://localhost:7687",
+    embedding_dimension=1024,   # must match your embedding model
+)
+
+# Qdrant — dedicated vector DB, cloud-ready
+v = Vektori(
+    storage_backend="qdrant",
+    database_url="http://localhost:6333",
+    embedding_dimension=1024,
+)
+
+# Qdrant Cloud
+v = Vektori(
+    storage_backend="qdrant",
+    database_url="https://your-cluster.qdrant.io",
+    qdrant_api_key="your-api-key",
+    embedding_dimension=1024,
+)
+
 # In-memory — tests / CI
 v = Vektori(storage_backend="memory")
 ```
 
-**Postgres via Docker:**
+**All backends via Docker:**
 ```bash
 git clone https://github.com/vektori-ai/vektori
 cd vektori
-docker compose up -d
+docker compose up -d                 # starts Postgres, Neo4j, and Qdrant
+
+# Postgres
 DATABASE_URL=postgresql://vektori:vektori@localhost:5432/vektori python examples/quickstart_postgres.py
+
+# Neo4j
+VEKTORI_STORAGE_BACKEND=neo4j VEKTORI_DATABASE_URL=bolt://localhost:7687 vektori add "I prefer dark mode" --user-id u1
+
+# Qdrant
+VEKTORI_STORAGE_BACKEND=qdrant VEKTORI_DATABASE_URL=http://localhost:6333 vektori add "I prefer dark mode" --user-id u1
+```
+
+**CLI storage flags:**
+```bash
+vektori config --storage-backend qdrant --database-url http://localhost:6333
+vektori add "my note" --user-id u1
+vektori search "preferences" --user-id u1
 ```
 
 ---

docker-compose.yml

@@ -17,5 +17,37 @@ services:
       timeout: 5s
       retries: 5
 
+  neo4j:
+    image: neo4j:5
+    environment:
+      NEO4J_AUTH: neo4j/password
+      NEO4J_PLUGINS: '["apoc"]'
+      NEO4J_dbms_security_procedures_unrestricted: apoc.*
+    ports:
+      - "7474:7474"   # HTTP browser
+      - "7687:7687"   # Bolt
+    volumes:
+      - neo4jdata:/data
+    healthcheck:
+      test: ["CMD", "cypher-shell", "-u", "neo4j", "-p", "password", "RETURN 1"]
+      interval: 10s
+      timeout: 10s
+      retries: 10
+
+  qdrant:
+    image: qdrant/qdrant:latest
+    ports:
+      - "6333:6333"   # REST
+      - "6334:6334"   # gRPC
+    volumes:
+      - qdrantdata:/qdrant/storage
+    healthcheck:
+      test: ["CMD-SHELL", "curl -sf http://localhost:6333/healthz || exit 1"]
+      interval: 5s
+      timeout: 5s
+      retries: 10
+
 volumes:
   pgdata:
+  neo4jdata:
+  qdrantdata: