Self-Managed Prometheus for Retina in Windows does not work

**Describe the bug**
self-managed Prometheus deployment does not work for retina windows pods.

**To Reproduce**
Steps to reproduce the behavior:
1. Deploy retina on windows
2. Go to (https://github.com/microsoft/retina/blob/main/docs/02-Installation/04-prometheus.md)
3. Follow the recommended steps
4. We will Retina targets are not resolved.
5. Metrics are not seen in Graphana

**Expected behavior**
Retina pod pool to be shown in the target with 10093 metrics endpoint recognized.

**Screenshots**
If applicable, add screenshots to help explain your problem.

this image shows that retina targets are not resolved.

**Platform (please complete the following information):**

- OS: Windows 2022
- Kubernetes Version: [e.g. 1.22]
- Retina Version: Latest

**Additional context**
The root cause:
1. The retina win pods require annotation - prometheus.io/scrape = true but it is not seen.
2. Here is how retain win daemon-set looks like --
kubectl describe daemonsets/retina-agent-win -n kube-system | findstr prometheus
prometheus.io/port: 10093
prometheus.io/scrape: true

kubectl describe pods/retina-agent-win-w2jdp -n kube-system | findstr prometheus

3. Scrape job looks for container name retina but retina windows pods look for retinawin

Start with docs/02-Installation/04-prometheus.md and the Prometheus scrape job that discovers Retina Windows pods; compare its container-name selector with retina-agent-win and retinawin and review the prometheus annotations. Reproduce on Windows 2022, then verify the Windows target resolves, recognizes endpoint 10093, and exposes metrics in Grafana.

Written by the indexing model from the issue text.