I am a beginner in optimization theory, So wait for someone to confirm this.
My idea is to start with finding the standard form of an SDP, witch is :
$$\begin{array}{rll} {\displaystyle\min_{X \in \mathbb{S}^n}} & \langle C, X \rangle_{\mathbb{S}^n} & \\ \text{subject to} & \langle A_i, X \rangle_{\mathbb{S}^n} = b_i, \quad i = 1,\ldots,m & (P) \\ & X \succeq 0 & \end{array}$$
To do that, I am thinking to use the same strategy as in my answer of this question I posed ( dual problem of a Semidefinite programming in a non-standard forme)
Let $ E^{(k, l)} \in S^ {9\times 9} $ ($k,l=1,..., 9)$ denote
the symmetric matrices with $(k, l)$th entry = $(l, k)$th entry $=
1$ and zero elsewhere. When $k=l$, $ E^{(k, k)} $ denotes the matrix
with ($k,k)$th entry 1 and all other elements 0. Clearly, we have
$E^{(l,k)}=E^{(k,l)}$ for any $(k,l).$ Note that for any matrix
$G=(g_{i,j}) \in S^{9\times 9},$ it can be represented as
$G=\sum_{k=1}^{9} \sum_{l=k}^{9} g_{k,l}E^{(k,l)},$ and $
\langle E^{(k, l)}, G \rangle = g_{k,l}+g_{l,k} = 2g_{k,l}$ for
$k\not=l,$ and $ \langle E^{(k, k)}, G \rangle = g_{k,k} $.
your problem can be written in the form :
$$
\begin{eqnarray} & \min & \left\langle
\frac{1}{2} W,
G \right\rangle \\
& \textrm{s.t.} & \langle 2E^{(1,1)}-E^{(1,i)}, G \rangle =0, ~i=2,...,
9, \\
& & \langle E^{(1,1)}-E^{(i, i)}, G \rangle =0, ~i=2,...,
9, \\
& & G \succeq 0
\end{eqnarray}
$$
The dual of this problem is well known.