[Feat]Take GraphAr as the data source for more graph processing systems

**Is your feature request related to a problem? Please describe.**
Currently, GraphAr is integrated into some existing systems as the data source to conduct graph processing, including GraphScope, BGL and GraphX. To validate usage scenarios of GraphAr, especially its libraries of different languages, more graph processing systems could be used to co-work with GraphAr.

**Describe the solution you'd like**
Consider existing graph processing systems of different languages (e.g., Apache Giraph, Apache Hama, Ligra, PowerGraph, etc), and integrate GraphAr into them as the data source.

**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.

**Additional context**
Add any other context or screenshots about the feature request here.

Start by reviewing GraphAr’s existing integrations with GraphScope, BGL, and GraphX, then compare candidate systems such as Apache Giraph, Apache Hama, Ligra, and PowerGraph. Done means choosing an additional graph processing system and integrating GraphAr as its data source, but the issue does not specify which system or an implementation entry point.

Written by the indexing model from the issue text.