1240 uploader look into processing h5ad in chunks

[adkinsrs]

(Still a WIP... need to test the rabbitMQ consumer)

This pull request removes legacy Python code related to dataset uploading and AnnData file handling, and adds documentation and service configuration for a new AnnData upload consumer. The changes help to streamline the codebase, reduce redundancy, and introduce a new, containerized RabbitMQ consumer for AnnData uploads.

Removal of legacy dataset upload code:

• Deleted lib/gear/dataarchive.py, which previously handled detection, parsing, and writing of various dataset formats (MEX, 3tab) into AnnData objects.
• Deleted lib/gear/datasetuploader.py, which included the dataset uploader factory and logic for determining file types and handling tarball contents.
• Deleted lib/gear/exceluploader.py, which implemented Excel file parsing and conversion to AnnData, including validation and statistics/coloring calculations.

Introduction of new AnnData upload consumer:

• Added a new anndata_upload_consumer service to the Docker Compose template (docker/docker-compose.yml.template), which builds from a dedicated Dockerfile and is responsible for handling AnnData and related file uploads via RabbitMQ.

Documentation updates:

• Updated the RabbitMQ consumers documentation to describe the new AnnData upload consumer, including its listener script, queue name, and systemd service details.