compute_sim_result: compute_sim_result
In rzgross/uRbanmatching:

Takes in functions to generate simulation data, and computes simulation results for our method vs. Mahalanobis and propensity score matching.

compute_sim_result(
  x_generator = default_x_generator,
  treat_prob_generator,
  mean_generator,
  error_generator = default_error_generator,
  n_sink_gen = n_sink_generator(),
  match_method = "with_replacement",
  n_rows = 500L,
  n_cols = 5L,
  num_weight_vectors = 100L,
  silent = !interactive()
)

`x_generator`	Function that takes number of rows and columns and produces a data matrix of that dimension.
`treat_prob_generator`	Function that takes a matrix and produces treatment probabilities.
`mean_generator`	Function that takes a matrix and produces an expected value for each row.
`error_generator`	Function that accepts a number of rows and generates an error vector (e.g. normal noise).
`n_sink_gen`	Default `n_sink_generator`. This argument should be a function that accepts a `treat_vec` and produces a vector of sink numbers.
`match_method`	"with_replacement", "optimal", or "greedy"
`n_rows`	How many rows to generate.
`n_cols`	How many columns to use.
`num_weight_vectors`	How many weight vectors to generate.
`silent`	Whether to suppress messages as it's running.. Default `!interactive()`.

Returns a named list:

naive_est: Mean difference between all treated units and all control
propensity_results: List of lists: each with n_sinks and the est
mahal_results: Same as above
weighted_results: List of lists: each with n_sinks, the raw brier score, the permutation brier score, and the est