I am asked to prove the comparison property for regulated functions, namely :
If $a \lt b$ and $f,g \in R[a,b]$ satisfy $f \leq g$ then $\int^b_af(x)dx \leq \int^b_ag(x)dx$.
The definition for regulated function is as follows:
Now, my proof is:
Let $\phi_n,\psi_n \in S[a,b]$ be sequences of step functions converging uniformly to $f$ and $g$ respectively and $\phi_n \leq \psi_n \forall x \in [a,b]$. Let $P=\{p_0,...,p_k\}$ be a partition compatible with both $\phi_n$ and $\psi_n$.
Then $\phi_n \leq \psi_n \Rightarrow f \leq g$ (I think $f \leq g \Rightarrow \phi_n \leq \psi_n$ and hence $\phi_n \leq \psi_n \Leftrightarrow f \leq g$, right?)
Furthermore, $\phi_n \leq \psi_n$
$\Leftrightarrow \sum^k_{i=1}\phi_n(p_i-p_{i-1}) \leq \sum^k_{i=1}\psi_n(p-p_{i-1})$
$\Leftrightarrow \text{lim}_{n \rightarrow \infty}\sum^k_{i=1}\phi_n(p_i-p_{i-1}) \leq \text{lim}_{n \rightarrow \infty} \sum^k_{i=1}\psi_n(p-p_{i-1})$
$\Leftrightarrow \int^b_a f(x)dx \leq \int^b_ag(x)dx$
Is my proof correct?

As pointed out in the comments, there is an issue because $f\le g$ does not imply $\phi_n \le \psi_n$. To correct it, let's introduce an $\epsilon>0$. Since $\phi_n$ and $\psi_n$ converge uniformly to $f$ and $g$, respectively, there exists $N\in \mathbb{N}$ such that $||f-\phi_n|| < \epsilon$ and $||g-\psi_n|| < \epsilon$ for all $n\ge N$ (where the norms here are the sup norm). Hence we have $f-\phi_n > -\epsilon$ and $g-\psi_n < \epsilon$, which can be rearranged to see that $\phi_n \le \psi_n + 2\epsilon$.
At this point, apply the same argument you used in your answer to obtain $$ \int_a^b f(x) dx \le \int_a^b g(x) dx + \int_a^b 2\epsilon dx = \int_a^b g(x) dx + 2(b-a)\epsilon $$ Since $\epsilon$ can be arbitrarily small, you can let it go to 0 and get the desired inequality.
In a nutshell: $f\le g$ doesn't necessarily imply $\phi_n\le \psi_n$, but it does imply $\phi_n$ is almost less than or equal to $\psi_n$.