Fix extremes n (#1413)

<!--Please ensure the PR fulfills the following requirements! -->
<!-- If this is your first PR, make sure to add your details to the
AUTHORS.rst! -->
### Pull Request Checklist:
- [ ] This PR addresses an already opened issue (for bug fixes /
features)
    - This PR fixes #xyz
- [ ] Tests for the changes have been added (for bug fixes / features)
- [ ] (If applicable) Documentation has been added / updated (for bug
fixes / features)
- [x] CHANGES.rst has been updated (with summary of main changes)
- [x] Link to issue (:issue:`number`) and pull request (:pull:`number`)
has been added

### What kind of change does this PR introduce?

*  add a bit more room in the guess of number of quantiles needed.

### Does this PR introduce a breaking change?


### Other information:
* Without the extra space, ExtremesValues fails on IPSL-CM6A-LR r3i1p1f1
when trained with EMDNA. I think this because IPSL has more extremes
than EMDNA on some blocks.

CHANGES.rst

@@ -4,7 +4,7 @@ Changelog
 
 v0.45.0 (unreleased)
 --------------------
-Contributors to this version: David Huard (:user:`huard`), Trevor James Smith (:user:`Zeitsperre`), Pascal Bourgault (:user:`aulemahal`).
+Contributors to this version: David Huard (:user:`huard`), Trevor James Smith (:user:`Zeitsperre`), Pascal Bourgault (:user:`aulemahal`), Juliette Lavoie (:user: `juliettelavoie`).
 
 New features and enhancements
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -18,6 +18,7 @@ Bug fixes
 Internal changes
 ^^^^^^^^^^^^^^^^
 * Tolerance thresholds for error in ``test_stats::test_fit`` have been relaxed to allow for more variation in the results. Previously untested ``*_moving_yearly_window`` functions are now tested. (:issue:`1400`, :pull:`1402`).
+* Increased the guess of number of quantiles needed in ExtremeValues. ( :pull:`1413`).
 
 v0.44.0 (2023-06-23)
 --------------------

xclim/sdba/adjustment.py

@@ -649,7 +649,7 @@ def _train(
         cluster_thresh = convert_units_to(cluster_thresh, ref, context="infer")
 
         # Approximation of how many "quantiles" values we will get:
-        N = (1 - q_thresh) * ref.time.size
+        N = (1 - q_thresh) * ref.time.size * 1.05  # extra padding for safety
 
         # ref_params: cast nan to f32 not to interfere with map_blocks dtype parsing
         #   ref and hist are f32, we want to have f32 in the output.