Design a social network with snapshots

You are asked to design and implement an in-memory **SocialNetwork** class that supports users following each other and creating **snapshots** of the follow graph. ### Requirements Implement an API similar to: ```text social_network = SocialNetwork() social_network.add_user("A") social_network.add_user("B") social_network.follow("A", "B") snapshot = social_network.create_snapshot() assert snapshot.is_follow("A", "B") ``` A **snapshot** represents an immutable view of the follow relationships at the moment it was created. ### Part 1 — Snapshot follow query - Support `snapshot.is_follow(u, v)` which returns whether user `u` follows user `v` **in that snapshot**. ### Part 2 — List followers and followees For a given snapshot and user: - Return the list/set of **followers** (who follows this user). - Return the list/set of **followees** (whom this user follows). ### Part 3 — Recommend users to follow Given a snapshot, a user `u`, and an integer `k`, recommend up to `k` users for `u` to follow using this logic: 1. Look at the users that `u` already follows. 2. For each of those followees, look at whom they follow. 3. Count how many times each candidate user is followed by `u`’s followees. 4. Return the top-`k` candidates by this count. ### Assumptions / clarifications (you may choose reasonable defaults) - What to do if `u` or `v` does not exist. - Whether to allow self-follow. - Whether repeated `follow` calls are idempotent. - Whether to exclude users that `u` already follows from recommendations. Describe your design (data structures, snapshot strategy, complexity) and any edge cases.