perf: optimize redundant allocations and loop operations

[justinbay]

This PR improves performance by removing redundant allocations and optimizing loop operations in the Cairo compiler codebase.

[reviewable-StarkWare]

This change is 

[orizi]

@orizi reviewed 1 of 2 files at r1, all commit messages.
Reviewable status: 1 of 2 files reviewed, 2 unresolved discussions

---

crates/cairo-lang-utils/src/graph_algos/strongly_connected_components.rs line 100 at r1 (raw file):

        };
    }
    // Update current_node in ctx.known_nodes after processing all neighbors.

this possibly changes the actual result, as it is recursive.

---

crates/cairo-lang-utils/src/graph_algos/strongly_connected_components.rs line 102 at r1 (raw file):

    // Update current_node in ctx.known_nodes after processing all neighbors.
    ctx.known_nodes.insert(current_node_id.clone(), current_wrapper_node.clone());
    

run ./scripts/rust_fmt.sh.