Event Timeline Reconstruction

You are given a log file that contains events, where each event is represented as a JSON object with the following structure:

{
  "id": "event1",
  "timestamp": "2025-01-04T12:00:00Z",
  "previous_event_id": null
}

Each event may depend on a previous event (indicated by the field previous_event_id). An event with a null value for previous_event_id starts a new chain. The events in the input array may be unordered and the chains can be independent.

Your task is to:

1. Parse the input log (an array of event objects).
2. Identify all chains of dependent events (i.e., where an event B depends on event A, event C depends on event B, etc.).
3. Reconstruct and return the longest chain of events as a list of event IDs (from the start of the chain to the end). If there are multiple chains of maximum length, you can return any one of them.

Make sure your solution can handle large datasets efficiently.