allow option to plot gridlines on pretty plot

docs/source/usage/getstart.rst

@@ -28,6 +28,7 @@ more about the powers of pre-commit from `their documentation <https://pre-commi
 
 For now, CLEO's python packages are not readily installable and so you will have to clone the CLEO
 repository to somewhere in your filesystem and checkout to the performance_testing_cleo branch:
+
 .. code-block:: console
 
   $ git clone https://github.com/yoctoyotta1024/CLEO.git

docs/source/usage/ourdocs.rst

@@ -18,7 +18,7 @@ which you can then view in your preferred browser e.g.
 
   $ open build/html/index.html
 
-Viewing Documentation Files on a Remote Machien Using your Local Browser
+Viewing Documentation Files on a Remote Machine Using your Local Browser
 ------------------------------------------------------------------------
 Let's say you're working on a remote machine (e.g. Levante) and have made some changes to the
 documentation in a branch of your repository. After you built the documentation via
@@ -35,13 +35,13 @@ not your local file system. Here is a good way to solve this problem exampled on
 
 #. SSH to Levante with local port forwarding enabled:
 
-    .. code-block:: console
+   .. code-block:: console
 
-      $ ssh -L 8765:localhost:8765 <userid>@levante.dkrz.de
+    $ ssh -L 8765:localhost:8765 <userid>@levante.dkrz.de
 
 #. Enter the directory than contains the ``index.html`` file and create an http server from it.
 
-    .. code-block:: console
+   .. code-block:: console
 
       $ cd docs/build/html/
       $ python -m http.server 8765

docs/source/usage/validations.rst

@@ -17,7 +17,7 @@ your executables ``main_[xxx].cpp`` file. E.g. uncomment the lines leading to
 ``return obssd >> obs1 >> obs0;`` and comment out the line ``return NullObserver{};``
 
 #. Then you could perform one normal run with e.g.
-``./builds/serial/condensation0d/cond0d ./builds/serial/tmp/cond0d/config_1_128_1_0.yaml``
+  ``./builds/serial/condensation0d/cond0d ./builds/serial/tmp/cond0d/config_1_128_1_0.yaml``
   (it is recomended that you create the output directory before running the model). Or
   you can run the example with Kokkos performance tool(s), e.g.
 

scripts/plotting/prettyplots_kerneltimer.py

@@ -8,7 +8,7 @@
 Author: Clara Bayley (CB)
 Additional Contributors:
 -----
-Last Modified: Thursday 27th March 2025
+Last Modified: Monday 1st September 2025
 Modified By: CB
 -----
 License: BSD 3-Clause "New" or "Revised" License
@@ -97,7 +97,10 @@ def perfect_scaling(x1, x2, y1, m=1):
 
 # %%
 def plot_wallclock_vs_total_num_supers(
-    datasets: dict, nthreads2plot: dict, simulated_time: float
+    datasets: dict,
+    nthreads2plot: dict,
+    simulated_time: float,
+    withlines: Optional[bool] = False,
 ):
     fig, axs = plt.subplots(nrows=2, ncols=2, figsize=(10.5, 7), sharex=True)
     axs = axs.flatten()
@@ -192,6 +195,10 @@ def plot_wallclock_vs_total_num_supers(
     for ax in axs:
         ax.set_ylim([5e-5, 5])
 
+    if withlines:
+        for ax in axs:
+            ax.grid(True)
+
     fig.tight_layout()
 
     return fig, axs
@@ -488,6 +495,12 @@ def plot_speedup_strong_scaling_for_total_num_supers(
 fig, axs = plot_wallclock_vs_total_num_supers(datasets, nthreads2plot, simulated_time)
 savename = path4plots / "wallclock_vs_totnsupers.png"
 save_figure(savename)
+
+fig, axs = plot_wallclock_vs_total_num_supers(
+    datasets, nthreads2plot, simulated_time, withlines=True
+)
+savename = path4plots / "wallclock_vs_totnsupers_withlines.png"
+save_figure(savename)
 # %%
 ngbxs2plot = {
     "Serial": [1, 64, 512, 4096, 32768],