Langroid

Neo4jChatAgent: better user interaction: with `interactive=False`, wait for user input when explicitly addressed via
`config.addressing_prefix`, or if LLM issues a non-tool message.

feat: add `OpenAIChatModel.GPT4o_MINI`

fix: neo4j_chat_agent agent_response should have sender=Entity.AGENT, not Entity.LLM. This was blocking LLM responses since LLM is not allowed to respond its own message

feat: `LanceDB`, schema-less ingest, compat with pydantic v2, (no need to create any explicit schema); `VectorStoreConfig` flexible document, metadata classes.

`VectorStoreConfig` has two new configs: `document_class` (any class derived from `Document`) and `metadata_class` (subclass of `DocMetadata`).
Specifying these classes allows more flexible document ingestion, see `test_vector_stores.py::test_vector_store_access` for examples.

feat: Recognize LLM-generated Tool/Fn-call when `request` field was forgotten but it is still uniquely recognizable via arguments specific to the Tool.

Fix: gpt4o on Azure, Momento, tool message parsing (525), qdrant docker mode

* gpt4o on Azure: handle their odd model version specification, see `handle_gpt4_model` fn in `azure_openai.py`
* chore: freeze momento vector-db version since new version breaks
* fix: `openai_gpt.py`: use `parse_imperfect_json()` for function arg parsing, to handle various edge cases.
* add docker mode for qdrant, see `QdrantDBConfig.docker`