check_predictions() for Bayesian models should use our own prediction/plotting capabilities

[DominiqueMakowski]

Somewhat old issue (#477, #706)

Instead of relying on bayesplot:

performance/R/check_predictions.R

Lines 217 to 225 in a6dcc28

	if (inherits(model, "brmsfit")) {
	out <- as.data.frame(
	bayesplot::pp_check(model, type = pp_type, ndraws = iterations, ...)$data
	)
	} else {
	out <- as.data.frame(
	bayesplot::pp_check(model, plotfun = pp_type, nreps = iterations, ...)$data
	)
	}

We could call modelbased::estimate_prediction(model_priors) and plot that using the same machinery as we do for other models

[copilot-swe-agent]

The agent encountered an error and was unable to start working on this issue: The service is temporarily unavailable. Please try again later. (Request id: 7E8A:36CAB5:4ECBF1:3B8405:69DD02F3)

[copilot-swe-agent]

The agent encountered an error and was unable to start working on this issue: This may be caused by a repository ruleset violation. See granting bypass permissions for the agent, or please contact support if the issue persists. (Request id: 3374:2FDDB6:838C5C:63210A:69DD066F)