refactor(mmm): rename optimize_budget **minimize_kwargs to **allocate_budget_kwargs

The variadic on BudgetOptimizerWrapper.optimize_budget relays every
keyword to BudgetOptimizer.allocate_budget, not only the scipy
minimize options, so the previous name was misleading (callers could
end up passing minimize_kwargs={...} inside **minimize_kwargs).

Closes #1660

Author: anevolbap

pymc_marketing/mmm/mmm.py

@@ -3681,7 +3681,7 @@ def optimize_budget(
         budget_distribution_over_period: xr.DataArray | None = None,
         cost_per_unit: pd.DataFrame | xr.DataArray | None = None,
         callback: bool = False,
-        **minimize_kwargs,
+        **allocate_budget_kwargs,
     ) -> (
         tuple[xr.DataArray, OptimizeResult]
         | tuple[xr.DataArray, OptimizeResult, list[dict[str, Any]]]
@@ -3732,8 +3732,10 @@ def optimize_budget(
             **This is independent of the historical cost_per_unit.**
         callback : bool
             Whether to return callback information tracking optimization progress.
-        **minimize_kwargs
-            Additional arguments for the optimizer.
+        **allocate_budget_kwargs
+            Extra keyword arguments forwarded to
+            :meth:`~pymc_marketing.mmm.budget_optimizer.BudgetOptimizer.allocate_budget`,
+            such as ``x0``, ``minimize_kwargs``, and ``return_if_fail``.
 
         Returns
         -------
@@ -3782,7 +3784,7 @@ def optimize_budget(
             total_budget=budget,
             budget_bounds=budget_bounds,
             callback=callback,
-            **minimize_kwargs,
+            **allocate_budget_kwargs,
         )
 
     def _apply_budget_distribution_pattern(