Add screenshots to blog 6.2: hero fix, dashboard metrics, swagger AI endpoints, hr-insight answer

Author: workcontrolgit

blogs/series-6-ai-app-features/6.2-dotnet-ai-hr-assistant.md

@@ -6,7 +6,7 @@ Generic chatbots answer generic questions. What makes an AI assistant useful in
 
 This article shows you how to build a data-aware HR assistant that pulls live workforce metrics from the database, injects them into the LLM prompt, and returns answers grounded in your actual data — not hallucinated statistics.
 
-![HR Insight page showing a data-grounded Ollama answer — references live department headcounts, execution time shown below reply](../../Tests/AngularNetTutorial-Playwright/screenshots-output/series-6-ai-app-features/ai-hr-insight-with-answer.png)
+![HR Insight page showing a data-grounded Ollama answer — references live department headcounts, execution time shown below reply](../../docs/screenshots/series-6-ai-app-features/ai-hr-insight-with-answer.png)
 
 📖 **Tutorial Repository:** [AngularNetTutorial on GitHub](https://github.com/workcontrolgit/AngularNetTutorial)
 
@@ -50,6 +50,8 @@ This is the classic **hallucination problem**: LLMs generate plausible-sounding
 
 For HR dashboard metrics — totals, distributions, recent hires — direct context injection is the right choice. The data is small (a few dozen lines of text), always current, and exactly what the LLM needs.
 
+![Dashboard showing KPI metric cards — total employees, departments, new hires, and Chart.js charts — the same data injected into the LLM system prompt](../../docs/screenshots/series-6-ai-app-features/dashboard-metrics-charts.png)
+
 ---
 
 ## 💡 The Solution
@@ -274,6 +276,8 @@ dotnet run
 
 Open Swagger at `https://localhost:44378/swagger` and find **POST /api/v1/ai/hr-insight**.
 
+![Swagger UI AI controller expanded — POST /ai/chat, POST /ai/hr-insight, POST /ai/nl-employee-search](../../docs/screenshots/series-6-ai-app-features/swagger-ai-endpoints.png)
+
 Try these questions:
 
 ```json
@@ -305,6 +309,8 @@ Try these questions:
 }
 ```
 
+![HR Insight page showing a data-grounded Ollama answer — references live department headcounts, execution time shown below reply](../../docs/screenshots/series-6-ai-app-features/ai-hr-insight-with-answer.png)
+
 **To test the feature flag** — set `"AiEnabled": false` in `appsettings.json` and restart the API. Both `/ai/chat` and `/ai/hr-insight` return `503 Service Unavailable`. No other endpoints are affected.
 
 ---