HSSM does not properly send priors to Bambi for an HSGP

[theonlydvr]

Describe the bug
When setting up a regression for a model parameter with an HSGP term, HSSM is unable to construct a model without raising the following error:

ValueError: The priors for an HSGP term must be passed within a dictionary. Keys must the names of the parameters of the covariance function and values are instances of bambi.Prior or numeric constants.

HSSM version
0.2.5

To Reproduce
Create an HSSM model as normal and provide a regression formula for any parameter like below:

 "name": "a",
 "formula": "a ~ 0 + hsgp(stim, m=10, c=2, centered=True)",
 "prior": {
     "hsgp(stim, m=10, c=2, centered=True)": {
          "name": "HSGP",
          "sigma": bmb.Prior("Exponential", lam=2), # amplitude
          "ell": bmb.Prior("InverseGamma", mu=10, sigma=1) # lengthscale
    },
    "sigma": Prior("HalfNormal", sigma=10)
}

Additional context
It seems like this issue occurs because HSSM automatically creates a bmb.Prior object while the HSGP expects to receive a dictionary.