[Genesis] update polling /ai-chat endpoint over the course of 90s (#434)

*Issue description:*
Main build for ADOT Python GenAi test case is flaky, failing most of the
time:
-
https://github.com/aws-observability/aws-otel-python-instrumentation/actions/runs/16299775127/job/46039999000

This is because metrics are usually not generated. We know that the
conditions to send metrics to CloudWatch is to generate a metric after
60s have passed since the last metric was generated.

Currently, we generate a metrics in a max 50 second interval, which
could possibly extend to 60s if the CURL request takes too long. So this
PR will change this to generate metrics over 90 seconds instead of 50
seconds to ensure we trigger sending metrics to CloudWatch.

*Description of changes:*

*Rollback procedure:*

We can safely revert this commit if needed.

Author: jj22ee

terraform/python/ec2/adot-genai/main.tf

@@ -113,7 +113,7 @@ done
 # Generate traffic directly
 echo "Starting traffic generator..."
 nohup bash -c '
-for i in {1..5}; do
+for i in {1..9}; do
     message="What is the weather like today?"
     echo "[$(date)] Request $i: $message"
     curl -s -X POST http://localhost:8000/ai-chat \