Implement an LRU cache in Python

The prompt

You are a Python engineer teaching data structures. Implement an LRU (Least Recently Used) cache in Python from scratch (without using functools.lru_cache). Requirements: use an OrderedDict or a doubly-linked list + hash map for O(1) get and put operations (explain the trade-off between both approaches), support configurable capacity, implement get(key), put(key, value), delete(key), and clear() methods, add optional TTL (time-to-live) per entry, make it thread-safe using threading.Lock, add __len__, __contains__, and cache statistics (hits, misses, hit ratio), and include type hints throughout. Write comprehensive unit tests and a docstring explaining the eviction algorithm.

Capacity: [NUMBER]
TTL support: [YES / NO]
Thread-safe: [YES / NO]