Add IPW diagnostics tests for fit parameter reporting

CHANGELOG.md

@@ -86,6 +86,8 @@
   - Added tests for `ascii_plot_dist` with `comparative=False` to verify direct dispatch to `ascii_plot_hist` and mixed categorical+numeric routing in a single call.
 - **Expanded `prop_above_and_below()` edge-case coverage**
   - Added focused tests for empty threshold iterables, mixed `None` threshold groups in dict mode, and explicit all-`None` threshold handling across return formats.
+- **Expanded IPW diagnostics coverage for fit-parameter reporting**
+  - Refactored diagnostics tests to use a shared IPW setup helper (removing repeated fixture construction) and added edge-case assertions for filtered non-string solver/penalty values and NaN coercion of non-scalar `tol`/`l1_ratio` inputs.
 
 # 0.16.0 (2026-02-09)
 

balance/sample_class.py

@@ -1894,7 +1894,6 @@ def diagnostics(
 
             fit_list: List[pd.DataFrame] = []
 
-            # TODO: add tests checking these values
             for array_key in ("n_iter_", "intercept_"):
                 array_val = getattr(fit, array_key, None)
                 if array_val is None:
@@ -1911,7 +1910,6 @@ def diagnostics(
                         )
                     )
 
-            # TODO: add tests checking these values
             for param_key, metric_name in (
                 ("penalty", "ipw_penalty"),
                 ("solver", "ipw_solver"),
@@ -1934,7 +1932,6 @@ def diagnostics(
                     )
                 )
 
-            # TODO: add tests checking these values
             multi_class = params.get("multi_class", getattr(fit, "multi_class", None))
             if multi_class is None:
                 multi_class = "auto"