UPDATE: after your comment I sat down and derived it myself (I couldn't make it out from the image), and I realised that your assumption that $\phi_0= \frac{\pi}{2}$ is not valid.
I prefer the following notation
$$x(t) = A e^{-\zeta \omega t } \sin\left(\sqrt{1-\zeta^2}\omega_n t +\phi_0\right)$$
and also substituting $\omega_d = \sqrt{1-\zeta^2}\omega_n$ (to keep equations shorter), this becomes:
$$x(t) = A e^{-\zeta \omega_n t } \sin\left(\omega_d t +\phi_0\right)$$
displacement BC
So for time $t=0$, $x(t=0)= X_0$.
therefore:
$$x(t=0) = A e^{0} \sin\left(\sqrt{1-\zeta^2}\cdot 0 +\phi_0\right)$$
$$x(t=0) = A \sin\left(\phi_0\right)$$
$$X_0 = A \sin\left(\phi_0\right)$$
velocity BC
By differentiating:
$$\dot{x}(t) = A \left((-\zeta \omega_n) e^{-\zeta \omega_n t } \sin\left(\omega_d t +\phi_0\right) + \omega_d e^{-\zeta \omega_n t } \cos\left(\omega_d t +\phi_0\right) \right)$$
Collecting term $ e^{-\zeta \omega_n t } $:
$$\dot{x}(t) = A e^{-\zeta \omega_n t } \left((-\zeta \omega_n) \sin\left(\omega_d t +\phi_0\right) + \omega_d \cos\left(\omega_d t +\phi_0\right) \right)$$
substituting and simplifying:
$$\dot{x}(t=0) = A \cdot 1 \cdot \left((-\zeta \omega_n)
\sin\left(\phi_0\right) + \omega_d \cos\left( \phi_0\right) \right)$$
$$0 = A \cdot 1 \cdot \left((-\zeta \omega_n) \sin\left(\phi_0\right) + \omega_d \cos\left( \phi_0\right) \right)$$
$$\zeta \omega_n \sin\left(\phi_0\right) = \omega_d \cos\left( \phi_0\right) $$
$$\frac{\sin\left(\phi_0\right)}{\cos\left( \phi_0\right)} = \frac{\omega_d}{\zeta \omega_n } $$
therefore
$$\tan\phi_0= \frac{\sqrt{1-\zeta^2}}{\zeta } $$
Interpretation
The two equations are
$$\begin{cases}
X_0 = A \sin\left(\phi_0\right)\\
\tan\phi_0= \frac{\sqrt{1-\zeta^2}}{\zeta }
\end{cases}$$
This is the actual correct solution. The assumption that $\phi_0=\frac{\pi}{2}$ is valid only for the case of the undamped free response. Whenever there is a damping ratio there is an angle $\phi_0$.
This can be observed in the following diagrams (I;ll get around to make them) for a $\zeta$ close to zero and close to 1.

The inflection point (change of curvature changes for different values of zeta) thus indicating the the phase angle changes.
In the case of
- $\zeta \rightarrow 0$ the initial change in displacement is almost horizontal (i.e. the velocity)
- $\zeta \rightarrow 1$ the initial change in displacement is very steep