Compute the eigenvalues of a 2×2 matrix, which describe how the matrix scales vectors along special directions called eigenvectors. You’ll implement a tiny helper that returns both eigenvalues in a consistent order.

The eigenvalues $\lambda$ and eigenvectors $v$ of a square matrix $A$ satisfy:

$$A v = \lambda v$$

The eigenvalues are found by solving the characteristic equation:

$$\det(A - \lambda I) = 0$$

The eigenvalues of

$$A=\begin{bmatrix}a & b\\ c & d\end{bmatrix}$$

are the roots of:

$$\lambda^2 - (a+d)\lambda + (ad-bc)=0$$

Requirements

Implement the function

Rules:

• Compute eigenvalues using the closed-form quadratic formula (not np.linalg.eig).
• Return a NumPy array of two floats, sorted from largest to smallest.
• If the eigenvalues are equal, return [lambda, lambda].
• Use only NumPy and Python built-in libraries.
• Keep the implementation in this single function.

Example

Output: