Define a composite SLO combining availability and latency using Datadog SLO API

Verified steps

1. Create the component SLOs first: an availability SLO (metric-based or monitor-based) and a latency SLO (metric-based using a Datadog histogram percentile metric); record the IDs of both created SLOs from the API response.
2. Create the composite SLO via POST to the /api/v1/slo endpoint with type set to 'composite'; supply the slo_ids array containing the IDs of the component SLOs and the composite rule, which is currently a logical AND (all component SLOs must be in-target for the composite to be in-target).
3. Set the target threshold and timeframe on the composite SLO; the composite SLO reports its status as the worst-performing component's error budget, i.e., the composite is breached if any component SLO is breached.
4. Retrieve the composite SLO status via GET /api/v1/slo/{slo_id}/history supplying from_ts and to_ts to get historical error budget data; the response includes overall_sli_value and error_budget fields.
5. Create a Datadog monitor using type slo alert and slo_id referencing the composite SLO ID to trigger alerts when the composite SLO's error budget falls below a threshold; configure separate thresholds for warn and critical states.
6. Add the composite SLO widget to a Datadog dashboard using the SLO widget type; it displays status, remaining error budget, and the contribution of each component SLO to the overall composite status.