C++: Fix broken lexing for floating-points

Now, I can have my Pi (3.14) typed as double.

Change-Id: I33ee579e56d3c735f88278f1868d8739ef277ad6
Reviewed-by: Orgad Shaneh <orgads@gmail.com>
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@qt.io>

src/libs/3rdparty/cplusplus/Literals.cpp

@@ -106,9 +106,11 @@ NumericLiteral::NumericLiteral(const char *chars, unsigned size)
             } // switch
         }
 
-        for (const char *dot = it; it != begin - 1; --it) {
-            if (*dot == '.')
+        for (const char *dot = it; dot != begin - 1; --dot) {
+            if (*dot == '.') {
                 f._type = NumericLiteralIsDouble;
+                break;
+            }
         }
 
         for (++it; it != end; ++it) {