Upgrade mem0ai to v1.0.0 with enhanced features

Major version upgrade from v0.1.118 to v1.0.0 bringing:
- Async mode as default for better performance
- Improved memory retrieval prompts
- v1.1 API format support (already configured in config.py)
- Better vector store integrations
- Enhanced LLM provider support
- Reranker support for multiple platforms

Updated dependencies:
- mem0ai[graph]: 0.1.118 → 1.0.0
- neo4j: 5.26.* → 5.28.2 (required by langchain-memgraph)
- openai: >=1.90.0,<1.110.0 → >=1.90.0 (no upper limit)

All tests passing with Ollama models (llama3.1:8b, nomic-embed-text).
Configuration already includes version: v1.1 for new API format.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

requirements.txt

@@ -1,6 +1,6 @@
 # Core Memory System
-# Requires >=0.1.118 for get_all() and search() dict return format fix
-mem0ai[graph]>=0.1.118,<0.2.0
+# v1.0.0 includes async by default, improved retrieval, and v1.1 API format
+mem0ai[graph]==1.0.0
 
 # Web Framework
 fastapi==0.115.*
@@ -13,12 +13,12 @@ mcp==1.3.*
 
 # Database Drivers
 psycopg2-binary==2.9.*
-neo4j==5.26.*
+neo4j>=5.28.1,<6.0.0
 vecs==0.4.*
 
 # OpenAI
-# mem0ai 0.1.118 requires openai<1.110.0,>=1.90.0
-openai>=1.90.0,<1.110.0
+# mem0ai 1.0.0 requires openai>=1.90.0 with no upper limit
+openai>=1.90.0
 
 # Ollama
 ollama>=0.4.0