I understand you're looking for a story involving a "Combinatorics and Graph Theory" solutions manual by Harris — likely referring to the textbook Combinatorics and Graph Theory by John M. Harris, Jeffry L. Hirst, and Michael J. Mossinghoff.
And at the very bottom of the acknowledgments, she wrote:
She wasn’t an instructor. She was a third-year Ph.D. student stuck on a single lemma about Hamiltonian cycles. But the basement had no security cameras, and her advisor had said, “Ask the library for miracles.” Combinatorics And Graph Theory Harris Solutions Manual
She saw the manual differently.
The solution was not a proof. It was a single diagram: a graph with 22 vertices and 33 edges, labeled like a constellation. At the bottom: This graph is you. Trace it. Find your odd cycle. I understand you're looking for a story involving
Problem 11.5: Construct a graph H such that the number of spanning trees of H is equal to the number of solutions to the Riemann Hypothesis with imaginary part less than 100.
Elena looked up from the manual and saw the library’s reading room not as a room, but as a graph . The desks were vertices. The students were edges — no, wait: students were walks between desks. She could see the adjacency matrix of the room pulsing faintly in the air. An undergrad shuffled past, and Elena instinctively computed: degree 3, not Eulerian, but close . Mossinghoff
She never told anyone where she’d found it.