Save results for last X queries into a memory buffer to revisit without re-querying

Some implementation of tracking the last X queries is already implemented.

For each of those queries, you could save the results as a .csv into a .folder, and circularly replace them when the X-sized buffer is at its end.

This way if you ran a 120 second query, followed by a 5 second query, I could revisit the results of the 120 second query without re-executing it.

No files, tests, or entry points are named. Start by locating the existing implementation that tracks recent queries, then determine how query results are represented and stored. Done means retaining results for the last X queries in CSV files, replacing the oldest entry circularly, and allowing an earlier result to be revisited without re-querying.

Written by the indexing model from the issue text.