chore: adjust evaulation periods for latency alarms (#314)

Update the evaluation periods for the latency alarms to only trigger if
the API has been consistently slow for 5 minutes.

This is to limit the number of false-positives being triggered by
API deploys.

terragrunt/aws/api/cloudwatch_api.tf

@@ -110,7 +110,7 @@ resource "aws_cloudwatch_metric_alarm" "api-gateway-above-maximum-latency-warnin
   alarm_name          = "api-gateway-above-maximum-latency-warning"
   alarm_description   = "API gateway latency between request and response above 1500ms"
   comparison_operator = "GreaterThanThreshold"
-  evaluation_periods  = "1"
+  evaluation_periods  = "5"
   metric_name         = "Latency"
   namespace           = "AWS/ApiGateway"
   period              = "60"
@@ -128,7 +128,7 @@ resource "aws_cloudwatch_metric_alarm" "api-gateway-above-maximum-latency-critic
   alarm_name          = "api-gateway-above-maximum-latency-critical"
   alarm_description   = "API gateway latency between request and response above 3000ms"
   comparison_operator = "GreaterThanThreshold"
-  evaluation_periods  = "1"
+  evaluation_periods  = "5"
   metric_name         = "Latency"
   namespace           = "AWS/ApiGateway"
   period              = "60"