Workaround Fan Percent Setting floating point err

Workaround floating point precision error which will cause invalid value
after ceil()

For example, the current value:
   speedMax: 10
   percent: 70
   speedMax * (percent * 0.01) = 7.000000000000001 (floating point
precision error)
   ceil(speedMax * (percent * 0.01)) = 8 => The error propagate to ceil
and cause the final result error.

src/app/clusters/fan-control-server/fan-control-server.cpp

@@ -347,7 +347,8 @@ void MatterFanControlClusterServerAttributeChangedCallback(const app::ConcreteAt
                            ChipLogError(Zcl, "Failed to get SpeedSetting with error: 0x%02x", status));
 
             float percent        = percentSetting.Value();
-            uint8_t speedSetting = static_cast<uint8_t>(ceil(speedMax * (percent * 0.01)));
+            // Minus insignificant number 0.00000001 before ceil() to avoid floating point precision error
+            uint8_t speedSetting = static_cast<uint8_t>(ceil(speedMax * (percent * 0.01) - 0.00000001));
 
             if (currentSpeedSetting.IsNull() || speedSetting != currentSpeedSetting.Value())
             {