Langroid

* Concurrency: Improvements to allow concurrent runs of `DocChatAgent`, see the
[`test_doc_chat_agent.py`](https://github.com/langroid/langroid/blob/main/tests/main/test_doc_chat_agent.py) in particular the `test_doc_chat_batch()`;
* Batch running of tasks: New task run utility: [`run_batch_task_gen`](https://github.com/langroid/langroid/blob/main/langroid/agent/batch.py)
where a task generator can be specified, to generate one task per input.

LanceDB nearest neighbor search - use `prefilter=True` when using a filter

- Adds a remote embeddings server for sentence transformer embeds, using grpc for remote procedure calls, allowing running langroid and generating local embeddings on different machines; currently, this functionality is not handled by tools such as ollama and litellm.

- If possible, uses data parallelism for local embeds; adds configurable batch sizes for all embeddings.

- Additionally, improves imports for parsing and vector stores.

run_batch_tasks` now takes a `turns` param to limit number of turns in each task (default = -1 meaning infinite)

ImagePdfParser`: extract text from image-based pdfs. This means `DocChatAgent` now works with image-based pdfs.

LanceDB: when creating schema, sort fields to prevent variable ordering and consequent mismatch between stored table and new docs during ingest.