Challenge 2: Fixing Challenge 1 With virtual — Possible Solution
====================================================================

#include <iostream>

class Shape {
public:
    virtual double area() { return 0; }
};

class Circle : public Shape {
public:
    double radius;
    double area() override { return 3.14159 * radius * radius; }
};

int main() {
    Shape *s = new Circle();
    static_cast<Circle*>(s)->radius = 5.0;

    std::cout << s->area() << std::endl;

    delete s;
    return 0;
}

Output:

    78.5397

Report: with virtual added to Shape::area() (and override on
Circle::area(), confirming it genuinely overrides the base), the
identical s->area() call through the same Shape* pointer now correctly
returns the circle's real area instead of 0. The result changed
because dispatch is now DYNAMIC -- resolved at runtime by looking up
the actual object's real type (Circle) through its vtable, rather than
statically by the pointer's own declared type (Shape). Nothing about
the call site (s->area()) changed at all between Challenge 1 and this
one -- only the virtual keyword did, and that single keyword is what
flips the dispatch mechanism entirely.

WHY THIS WORKS AS AN ANSWER
------------------------------
This isolates the ONE change (adding virtual/override) and shows the
call site is otherwise identical, correctly attributing the changed
result specifically to dispatch now happening at runtime via the
object's real type rather than the pointer's declared type -- the
precise mechanism the chapter names.