## Problem: Autocomplete suggestions (Trie) You are building an autocomplete feature. ### Input - A dictionary of `n` unique words `words[i]` (lowercase a–z). - An integer array `freq[i]` where `freq[i]` is the usage count of `words[i]`. - Then `q` queries. Each query provides: - a `prefix` string - an integer `k` (number of suggestions requested) ### Output For each query, return up to `k` suggested completions: - A completion is any word that starts with `prefix`. - Rank by **higher frequency first**. - Break ties by **lexicographical order**. ### Follow-up Support updates: - Operation `add(word, delta)` increases that word’s frequency by `delta` (word may be new). - Queries after updates must reflect the new frequencies. ### Constraints (reasonable interview assumptions) - `1 ≤ n, q ≤ 2e5` - `1 ≤ |word|, |prefix| ≤ 50` - Total characters across all words/queries can be large; solutions should be better than scanning all words per query. Explain what data structure you would use (e.g., Trie) and implement the required operations.