Design a voting counter for restaurants

You are building an in-memory voting component for a lunch poll. ## Task Design and implement a hash-table–backed data structure that supports collecting votes for restaurant names and reporting the most popular restaurant(s). ## Requirements - Each vote is a string `restaurantName`. - You must maintain vote counts as votes arrive. - Support these operations: 1. `vote(restaurantName)`: record one vote for the restaurant. 2. `getCount(restaurantName) -> int`: return the current vote count (0 if never seen). 3. `getWinner() -> string`: return the restaurant with the highest vote count. - If there is a tie, return the lexicographically smallest name (or clearly state and implement a deterministic tie-break rule). ## Constraints (assume) - Up to `N` total votes, where `N` can be large (e.g., 10^5–10^6). - Restaurant names can repeat and may include spaces. - Aim for near O(1) average time per `vote`/`getCount` using hashing. ## What to discuss - Data structures used and why. - Time and space complexity. - Edge cases: no votes yet, ties, unseen restaurant names.