annslicer

Create a utility named 'AnnDataShardMaster' that simplifies the process of managing large-scale biological datasets stored in .h5ad AnnData files using the 'annslicer' package. This utility will enable researchers to efficiently slice and shard their data into smaller, manageable parts without consuming excessive memory resources. The application should include the following key functionalities:

1. **Data Sharding**: Implement a feature that allows users to input a path to a large .h5ad file and specify the desired size of each shard. The utility should then automatically split the dataset into multiple smaller files based on the specified shard size.
2. **Out-of-Core Processing**: Ensure that the utility leverages 'annslicer' to perform out-of-core operations, meaning it can handle datasets larger than the available RAM. Users should be able to process these shards sequentially or concurrently as needed.
3. **Memory Management**: Integrate monitoring tools within the utility to track memory usage during the sharding process. This helps in understanding the efficiency of 'annslicer' in minimizing memory consumption.
4. **Interactive Interface**: Develop a simple command-line interface (CLI) where users can interact with the utility, providing inputs like file paths, shard sizes, and processing options.
5. **Documentation and Examples**: Provide comprehensive documentation along with sample datasets and use cases to help new users understand how to utilize 'AnnDataShardMaster'.

To achieve these objectives, make sure to utilize the core functionalities of the 'annslicer' package such as its ability to efficiently manage large datasets and perform operations without loading entire datasets into memory. Additionally, consider adding advanced features like automatic compression of shards, support for parallel processing, and error handling mechanisms to enhance the robustness of the utility.