Per-User Minimum Time Difference Between Orders and Ads Events (Two Sorted Lists)

You are given two lists sorted ascending by timestamp: orders: user order events ads: user ads-related events Each event contains at least userId and an integer timestamp. For each userId, find a pair of events (orderEvent, adEvent)—one from orders and one from ads for that same user—such that the absolute time difference |orderTimestamp - adTimestamp| is minimized. Return the minimum time difference for each user (optionally also the corresponding pair). Constraints/notes: Both input lists are globally sorted by timestamp (events of different users may be interleaved). If a user appears in only one list, return no result for that user. Assume: len(orders), len(ads) up to 1e5 timestamp fits in 64-bit integer Implement with the best possible time complexity. Example Input orders = [(u1,1),(u1,10),(u2,5)] ads = [(u1,6),(u2,100)] Output u1 -> 4 (order 10, ad 6) u2 -> 95 (order 5, ad 100)