1) What is the nontrivial 1-dimensional sub-manifold generator $M^1$ in $\mathbb{RP}^5$ of $H^1(\mathbb{RP}^5,\mathbb{Z}_2)$? How to visualize it where this $M^1$ sits in $\mathbb{RP}^5$?
2) What is the nontrivial 4-dimensional Poincaré dual sub-manifold generator PD$(M^1)$ for $H^1(\mathbb{RP}^5,\mathbb{Z}_2)$ in $\mathbb{RP}^5$? How to visualize it where this PD$(M^1)$ sits in $\mathbb{RP}^5$?
My attempt: I think the answer could be just a complex projective space PD$(M^1)$=$\mathbb{CP}^2$. But I hope to see more evidence or an intuitive proof for it.
I checked that all the Stiefel Whitney classes of $\mathbb{RP}^5$ and $\mathbb{CP}^2$ are the same.
As written, your questions don't quite make sense. I have tried to clarify below.
If $M$ is a closed $k$-manifold, then there is a homology class $[M] \in H_k(M; \mathbb{Z}_2)$ called the fundamental class of $M$. Given a continuous map $f : M \to N$, then $f_*[M] \in H_k(N; \mathbb{Z}_2)$. In particular, if $M$ is a closed submanifold of $N$, and $i : M \to N$ is the inclusion, then $i_*[M] \in H_k(N; \mathbb{Z}_2)$. We say a homology class $a \in H_k(N; \mathbb{Z}_2)$ can be represented by a submanifold $M$ if $i_*[M] = a$.
Given a cohomology class, you could ask that its Poincaré dual is represented by a submanifold. In particular, for the non-zero element $x \in H^1(\mathbb{RP}^5; \mathbb{Z}_2)$, you could ask whether $\operatorname{PD}(x) \in H_4(\mathbb{RP}^5; \mathbb{Z}_2)$ could be represented by a (four-dimensional) submanifold $M$. The answer turns out to be yes, with $M = \mathbb{RP}^4$. One possible embedding $\mathbb{RP}^4 \to \mathbb{RP}^5$ is given by $[x_0, \dots, x_4] \mapsto [x_0, \dots, x_4, 0]$.
Two different spaces can't have the same Stiefel-Whitney classes as they live in their respective cohomology groups. While $w_2(\mathbb{RP}^5) \in H^2(\mathbb{RP}^5; \mathbb{Z}_2) \cong \mathbb{Z}_2$ and $w_2(\mathbb{CP}^2) \in H^2(\mathbb{CP}^2; \mathbb{Z}_2) \cong \mathbb{Z}_2$ are both non-zero, they are not equal.