Here is a better example, showing that not only the answer to the OP is positive, but it is even possible to enforce on $f$ some additional, and fairly strong, conditions.
To start with, let $\theta$ be the function $\mathcal P(\mathbf N^+) \to \mathbf R$ defined as follows: Given $X \subseteq {\bf N}^+$, we assume $\theta(X) := 0$ if $|X| < \infty$, otherwise $\theta(X) := \sup_{n \ge 1} (x_{n+1} - x_n)^{-1}$, where $(x_n)_{n \ge 1}$ is the natural enumeration of $X$. Next, let $f$ be the function
$$
\mathcal P(\mathbf N^+) \to \mathbf R: X \mapsto \sqrt{\theta(X)\, {\sf d}^\ast(X)},
$$
where ${\sf d}^\ast$ is the upper asymptotic density on $\mathbf N^+$.
It is seen that $f$ is monotone and affinely $(-1)$-homogeneous, essentially because the same is true of $\theta$ and ${\sf d}^\ast$, see the note below; moreover, $f(\emptyset) = 0$ and $f(\mathbf N^+) = 1$.
We want to show that $f$ has the weak Darboux property. For this, fix $X \subseteq \mathbf N^+$ with $f(X) > 0$. Then $\mathsf{d}^\ast(X) > 0$, which implies that $|X| = \infty$ and there exist $x,y \in X$ such that $x < y$ and $f(X) = (y-x)^{-\frac{1}{2}}\sqrt{\mathsf{d}^\ast(X)}$.
Accordingly, let $a \in {]0, f(X)]}$, so that $0 \le (y-x) a^2 \le \mathsf{d}^\ast(X)$.
Since $\mathsf{d}^\ast$ has the weak Darboux property, see [1], we can find $Y \subseteq X$ such that $\mathsf{d}^\ast(Y) = (y-x) a^2$; set $A := Y \cup \{x,y\}$. It then follows from the monotonicity of $\theta$ that $\theta(A) = \theta(X) = (y-x)^{-1}$, while it is clear that $\mathsf{d}^\ast(A) = \mathsf{d}^\ast(Y)$. Thus, it is seen that $f(A) = a$, which ultimately shows that $f$ has the weak Darboux property.
Now, let $X := 4 \cdot \mathbf N^+$ and $Y := X \cup \{2\}$. It is straightforward that $\theta(X) = \mathsf{d}^\ast(X) = \mathsf{d}^\ast(Y) = \frac{1}{4}$ and $\theta(Y) = \frac{1}{2}$, with the result that $f(X) = \frac{1}{4}$ and $f(Y) = \frac{1}{4}\sqrt{2}$. Therefore, $f$ does not have the strong Darboux property, because $f(X) < f(Y)$, but $Y \setminus X$ is finite.
Notes. Here, a function $g: \mathcal P(\mathbf N^+) \to \mathbf R$ is said to be affinely $(-1)$-homogeneous if $g(k \cdot X + h) = g(X)$ for all $X \subseteq \mathbf N^+$ and $h, k \in \mathbf N^+$, where $k \cdot X + h := \{kx+h: x \in X\}$.
Bibliography.
[1] G. Grekos, Répartition des densités des sous-suites d'une suite d'entiers, J. Number Theory 10 (1978), No. 2, 177-191