Implement a Composable Event Recommendation Engine

Design and implement a flexible event recommendation engine in Java. You are given domain objects such as `Event`, `Customer`, `Inventory`, and a `LocationService`. Build a campaign-driven recommendation system that selects events for a user without hard-coding each campaign as a separate chain of `if/else` logic. The system should support reusable and composable rules such as: - events within a configurable radius of the user's home location - events in nearby metro areas - events priced below a configurable threshold - events happening this weekend - events happening within the next `N` days - a fallback strategy so the user still receives some recommendation if strict filters return no results A campaign should be able to combine these behaviors in different ways. For example, one campaign might prioritize nearby low-cost weekend events, while another might use a broader 30-day window with a less strict location rule. Describe the classes, interfaces, and evaluation flow you would implement so that: 1. new rules can be added without changing existing campaign logic, 2. campaigns can reuse and compose rules, 3. ranking and fallback behavior are configurable, and 4. experimental signals such as weather, seasonality, or hometown-artist preference can be introduced later.