O(1) LRU cache with correct eviction

Results are ordered by this task's published score. API runs use one prompt and one completion; agent runs use a coding harness with tools and iteration. Inspect the execution method when comparing results.

Objective scores use the benchmark's execution checks. Judge scores use a vision panel's fixed rubric; the published score leaves out a model's own family. Missing values are labelled rather than treated as zero.

Implement a Python class `LRUCache` (a least-recently-used cache) with these methods, each running in O(1) average time: - `LRUCache(capacity)`: construct an empty cache that holds at most `capacity` keys. `capacity` is a non-negative integer. - `get(key)`: return the value for `key`, or -1 if the key is not present. A successful get marks the key as the most recently used. - `put(key, value)`: insert or update the value for `key`, and mark it as the most recently used. If this makes the cache exceed `capacity`, evict the least recently used key first. Edge cases that matter: - A `get` counts as a use: it must refresh the key's recency. - Updating an existing key with `put` also refreshes its recency and must not grow the cache. - `capacity == 0` stores nothing: every `put` is a no-op and `get` always returns -1. Use only the Python standard library. Write your solution to `solution.py`.