[Feat]: Add Alerting support

[AmanAgarwal041]

🚀 What's the Problem?

In our current setup with the Doku, and monitoring OpenAI API Calls, we lack the ability to proactively monitor and react to critical metrics such as usage spikes, approaching request limit thresholds, and other relevant statistics that are pivotal for maintaining operational stability and cost management. Without real-time alerts, teams are often reactive rather than proactive, leading to disrupted services or unnecessary expenses.

💡 Your Dream Solution

A robust, configurable alerting system within LLM Observability that allows users to:

1. **Pre-built Alerts:**Alerts should be prebuilt into Doku with proper default thresholds
2. Set Custom Thresholds: Users should be able to define custom thresholds for various metrics (e.g., number of requests, usage spikes) beyond which alerts would be triggered.
3. Multi-Channel Alerts: Alerts should be dispatchable through multiple channels, including email, SMS, Slack, or webhooks, to ensure prompt notifications. We can start with email and Slack maybe?
4. Comprehensive Dashboard: An integrated dashboard that displays current metrics, historical data, and alert configuration settings for easy monitoring and management.

🤔 Seen anything similar?

Similar to Prometheus AlertManager where user can silence alerts, send to different channels etc.

🖼️ Pictures or Drawings

Share any picture or drawing that shows your idea if you have.

👐 Want to Help Make It Happen?

• Yes, I'd like to volunteer and help out with this!