The Irreducible Representations of the Sekine Quantum Groups Here Y. Sekine introduces a one-parameter family of finite quantum groups of dimension $2n^2$. Let $n\geq 3$ be fixed and $\zeta=e^{2\pi i/n}$. Set
$$\mathcal{B}_n=\mathbb{Z}_n\times\mathbb{Z}_n=\{(i,j):i,j=0,1,\dots,n-1\},$$
 and define a block matrix (of unitaries)  $U\in M_n(M_n(\mathbb{C}))$ by:
 $$u_{ij}=\sum_{m=1}^n\zeta^{im}E_{m,m+j},$$
 where $E_{i,j}$ is the usual matrix unit in $M_n(\mathbb{C})$ and $m+j$ is understood as $\mod n$.  Note the indexing of the  blocks is from $0\rightarrow k-1$ rather than $1\rightarrow k$ although the indexing inside the blocks is as standard.
Consider $n^2$ one-dimensional spaces $\mathbb{C} e_{(i,j)}$ spanned by elements indexed by $\mathcal{B}_n$, $\{e_{(i,j)}:i,j\in \mathcal{B}_n\}$. Together with a copy of $M_n(\mathbb{C})$, a direct sum of these $n^2+1$ spaces, the $2n^2$ dimensional space
 $$A_n=\left(\bigoplus_{(i,j)\in\mathcal{B}_n}\mathbb{C} e_{(i,j)}\right)\oplus M_n(\mathbb{C}),$$
 can be given the structure of the algebra of functions on a finite group which is denoted by $\mathbb{KP}_n$ (so that $A_n=F(\mathbb{KP}_n)$). On the one dimensional elements the coproduct is given by:
 $$\Delta(e_{(i,j)})=\sum_{(m,n)\in\mathcal{B}_n}e_{(m,n)}\otimes e_{(i-m,j-n)}+\frac{1}{n}\sum_{m,n,s,t=1}^n\left(u_{(i,j)}\right)_{m,s}\overline{\left(u_{(i,j)}\right)_{n,t}}e_{mn}\otimes e_{st}.$$
 On elements in the $M_n(\mathbb{C})$ factor:
 $$\Delta(a)=\sum_{(i,j)\in\mathcal{B}_n}e_{(-i,-j)}\otimes u_{(i,j)}a u_{(i,j)}^*+\sum_{(i,j)\in\mathcal{B}_n}\overline{u_{(i,j)}}au_{(i,j)}^T\otimes e_{(i,j)}.$$
The antipode is given by $S(e_{(i,j)})=e_{(-i,-j)}$ on the one dimensional factors and the transpose for the $M_n(\mathbb{C})$ factor. Sekine doesn't give the counit but by noting that $u_{(0,0)}=I_{n}$ it can be seen  that the projection onto the $e_{(0,0)}$ one-dimensional factor satisfies the counital property. The Haar state $h_n\in M_p(\mathbb{KP}_n)$ is given by:
 $$h_n\left(\sum_{(i,j)\in \mathcal{B}_n}x_{(i,j)}e_{(i,j)}+a\right)=\frac{1}{2n^2}\left(\sum_{(i,j)\in\mathcal{B}_n}x_{(i,j)}+n\cdot \text{Tr}(a)\right).$$

I am looking for a reference where the (co)representation theory is
  discussed. Such a reference may not exist --- Google isn't showing
  anything for me. I am also interested in the states.

In this question the quantum group is self-dual but this isn't the case as far as I know for these quantum groups. The linked paper above describes central projections in 'a' convolution algebra...
 A: Disclamer: This is just what some digging gave me, maybe some of it can be helpful:
Representation Theory:
The paper on nilpotent fusion categories [1] (see example 4.5(2)) claims that representations over the Kac-Paljutkin algebras (or even Sekine's generalizations) give examples of nilpotent semisimple tensor categories. In particular, it is claimed that these categories are examples of Tambara-Yamagami categories (introduced in [2], see e.g. [3] for a definition).
States:
About the states, there is a paper by Franz and Skalski [4] (section 6, Lemma 6.4) investigates all quantum subgroups and examples of idempotent states that are not Haar states on subgroups.
[1] Gelaki, Shlomo; Nikshych, Dmitri. Nilpotent fusion categories. Adv. Math. 217 (2008), no. 3, 1053--1071. 
[2] Tambara, Daisuke; Yamagami, Shigeru. Tensor categories with fusion rules of self-duality for finite abelian groups. J. Algebra 209 (1998), no. 2, 692--707.
[3] Turaev, Vladimir; Vainerman, Leonid. The Tambara-Yamagami categories and 3-manifold invariants. Enseign. Math. (2) 58 (2012), no. 1-2, 131--146. 
[4] Franz, Uwe; Skalski, Adam. On idempotent states on quantum groups. J. Algebra 322 (2009), no. 5, 1774--1802.
A: Thank you to Zahlendreher and Sébastien Palcoux for the help.
Zahlendereher led me to the states 
With Sébastien's help I was happy that brute force would reveal the matrix elements of the corepresentations to be related to the central minimal projections in the convolution algebra. I was unable to adapt his exact methods as our definitions were at odds but from what he told me I knew brute force wasn't futile.
I was able to see that for $n$ odd at least (all I need at this time --- I suggest that things are similar for $n$ even), the one-dimensional central minimal projections of Lemma 4 in the convolution algebra coincide with the matrix elements of the one-dimensional representations. 
Dividing the matrix units in the range of the two-dimensional central minimal projections of Lemma 5 by two gave the matrix elements of the two-dimensional representations.
