Function parallel_pagerank 

pub fn parallel_pagerank(
    graph: &CsrGraph,
    config: &PageRankConfig,
) -> Result<PageRankResult>

Compute PageRank using parallel power iteration.

At each iteration:

1. Compute the stochastic matrix column contributions in parallel
2. Apply the PageRank formula: r = (1-d)/n + d * A_norm * r_old
3. Check L1 convergence

Dangling nodes (out-degree 0) distribute their rank equally to all nodes.

Arguments

• graph - A directed CSR graph (or undirected treated as bidirectional)
• config - PageRank parameters

Returns

A PageRankResult with scores and convergence information.