Added a prompt injection detection (#857)

Co-authored-by: Thorsten Sommer <SommerEngineering@users.noreply.github.com>

app/MindWork AI Studio/Tools/Rust/SanitizePromptInjectionsRequest.cs

@@ -0,0 +1,6 @@
+using System.Text.Json.Serialization;
+
+namespace AIStudio.Tools.Rust;
+
+/// <param name="Text">The content to filter.</param>
+public readonly record struct SanitizePromptInjectionsRequest([property: JsonPropertyName("text")] string Text);

app/MindWork AI Studio/Tools/Rust/SanitizePromptInjectionsResponse.cs

@@ -0,0 +1,13 @@
+using System.Text.Json.Serialization;
+
+using AIStudio.Tools.Security;
+
+namespace AIStudio.Tools.Rust;
+
+/// <param name="SanitizedText">The content with the suspicious passages removed. Usable as it stands.</param>
+/// <param name="Findings">The passages that were removed, capped by the runtime.</param>
+/// <param name="RedactedCount">How many passages were removed in total, which may exceed the number of findings.</param>
+public readonly record struct SanitizePromptInjectionsResponse(
+    [property: JsonPropertyName("sanitized_text")] string SanitizedText,
+    [property: JsonPropertyName("findings")] IReadOnlyList<PromptInjectionFinding> Findings,
+    [property: JsonPropertyName("redacted_count")] int RedactedCount);