Add Kahan-Babuška Summation Algorithm and Enhance Variance Indicator Tests

- Introduced a new `Sum` class implementing the Kahan-Babuška algorithm for high-precision rolling summation.
- Added comprehensive documentation for the `Sum` class, detailing its mathematical foundation, performance profile, and use cases.
- Refactored `VarianceIndicator` tests to improve clarity and coverage, including checks for different source types and the ability to change properties.
- Enhanced `UsfIndicator` tests to validate initialization, processing of updates, and property changes.
- Updated `UsfIndicator` implementation to simplify source handling and improve short name generation.
- Modified Qodana configuration to exclude unused auto property accessor warnings.

lib/statistics/stddev/StdDev.Tests.cs

@@ -176,7 +176,7 @@ public class StdDevTests
     [Fact]
     public void IsHot_BecomesTrueAfterPeriod()
     {
-        int period = 5;
+        const int period = 5;
         var stdDev = new StdDev(period);
 
         for (int i = 0; i < period; i++)