[Blocker] Add health checks, telemetry and structured logging

Tier 1 — release blocker

Problem

There is no /healthz endpoint, no Application Insights, no OpenTelemetry and no structured logging. When production misbehaves the only signal is console output. Blazor Server in particular needs circuit-level visibility — dropped circuits, unhandled component exceptions and SignalR reconnects are invisible today.

Evidence

• No AddHealthChecks, ApplicationInsights, OpenTelemetry or Serilog reference anywhere
• LearnStack/Program.cs — default console logging only

Proposed fix

1. AddHealthChecks().AddDbContextCheck<ApplicationDbContext>() mapped at /healthz, wired to the App Service health probe.
2. Add Application Insights (or OpenTelemetry exporting to Azure Monitor).
3. Track: circuit open/close, unhandled component exceptions, OpenGraph fetch failures, login failures, slow queries.
4. Add a dashboard and an alert on the error rate.

Acceptance criteria

• /healthz returns degraded when the database is unreachable
• Unhandled exceptions in a Blazor circuit appear in telemetry with user/session correlation
• An alert fires on sustained 5xx or circuit-failure rate

Start by reading LearnStack/Program.cs to understand the current hosting and logging setup, then locate the database context and App Service configuration. The work is done when /healthz reports database availability, Blazor circuit failures and specified application failures are correlated in telemetry, and the requested error-rate alert and dashboard are configured.