Merged PR 3731: Add 1e-6 threshold in kpis_additive_metrics_per_dimension_are_consistent

# Description

Add 1e-6 threshold in kpis_additive_metrics_per_dimension_are_consistent. This should fix the issue on churn rate metrics that are rounded to this precision

# Checklist

- [X] The edited models and dependants run properly with production data. Only tried the test.
- [ ] The edited models are sufficiently documented.
- [X] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.

# Other

- [ ] Check if a full-refresh is required after this PR is merged.

Related work items: #25075

tests/kpis_additive_metrics_per_dimension_are_consistent.sql

@@ -4,13 +4,16 @@ specifically on reporting.mtd_aggregated_metrics.
 
 It just ensures that for the last available date, the sum of metrics
 for any dimension provides an equal or lower aggregated value compared
-to what is expected at Global level. This is because some dimensions
-depend on Deal, and not all users have a Deal, thus it's normal that
-the aggregation might not match the Global value on a given metric. 
+to what is expected at Global level, up to a certain threshold. 
+This is because some dimensions depend on Deal, and not all users 
+have a Deal, thus it's normal that the aggregation might not match the 
+Global value on a given metric. 
 
-However, the aggregation cannot be higher than the value reported in
-the Global dimension.
+However, the aggregation cannot be higher, with a certain tolerance, 
+than the value reported in the Global dimension.
 */
+{% set tolerance_threshold = 0.000001 %}
+
 {% set additive_metric_names = (
     "Cancelled Bookings",
     "Check-In Hero Revenue",
@@ -91,4 +94,4 @@ with
     )
 select *
 from difference_computation
-where abs_diff > 0
+where abs_diff > {{ tolerance_threshold }}