sim_rand_bandit: Simulate Random Bandit
In dfcorbin/npbanditC: (Contextual) Multi-Armed Bandit Simulations

In this function we assume that the true expectation of each lever can be expressed as linear combination of known basis functions with known coefficients.

sim_rand_bandit(
  nsteps,
  nlevers,
  J_mod,
  J_true,
  sd,
  alpha = 1,
  alpha_true = 1,
  calc_mse = FALSE,
  b = 2
)

`nsteps`	The number of steps to make in the trajectory.
`nlevers`	The number of levers to choose from.
`J_mod`	The number of basis functions to use in the model (not including) the intercept.
`J_true`	The number of basis functions to be used in the expression for each lever's true expectation.
`sd`	A vector of the standard deviations of each lever.
`alpha`	A scaling parameter which tunes how the prior variance decreases as J_mod increases.
`alpha_true`	The true scaling parameter that the coefficients are generated from.
`calc_mse`	True, if you want to output the mse for the beta coefficients.
`b`	Hyperparameter for the inverse gamma distribution. Directly related to the rate of exploration.