Example Request - Sending metrics to Amazon Managed Prometheus

**Describe the tutorial request**

My customer has been struggling to get metrics from AgentCore delivered to (Amazon Managed) Prometheus via OpenTelemetry, instead of CloudWatch - so would like to suggest we add an example of that alongside our other observability partner tutorials 🙏

Going by the [Langfuse](https://github.com/awslabs/amazon-bedrock-agentcore-samples/blob/main/01-tutorials/06-AgentCore-observability/04-Agentcore-runtime-partner-observability/Langfuse/runtime_with_strands_and_langfuse.ipynb) and [Braintrust](https://github.com/awslabs/amazon-bedrock-agentcore-samples/blob/897ed72cac5447954c8f2455f15d7142d07465ea/01-tutorials/06-AgentCore-observability/04-Agentcore-runtime-partner-observability/Braintrust/runtime_with_strands_and_braintrust.ipynb#L153) examples, I believe overriding AgentCore's pre-initialized OTel environment variables might be key to success - but haven't had chance yet to dig in and create a working example yet myself.

**Is this feature already supported by the AgentCore service?**

- [ ] Yes
- [ ] No
- [X] Not sure

I think it *should* be, as an OTel-based destination, but haven't found explicit mentions of AMP in the docs yet?