applications: machine_learning: Add power optimized configuration

Add power optimized configuration (S2RAM) to machine_learning
application.

Jira: NCSDK-32067

Signed-off-by: Dominik Chat <[email protected]>

Author: dchat-nordic

applications/machine_learning/sample.yaml

@@ -93,3 +93,20 @@ tests:
       - machine_learning_SNIPPET=nordic-ppr
       - machine_learning_SHIELD=pca63566_fwd
       - remote_SHIELD=pca63566
+  applications.machine_learning.sensor_hub.zdebug.singlecore.power_consumption_test:
+    build_only: true
+    sysbuild: true
+    platform_allow:
+      - nrf54h20dk/nrf54h20/cpuapp
+    integration_platforms:
+      - nrf54h20dk/nrf54h20/cpuapp
+    tags:
+      - ci_build
+      - sysbuild
+      - ci_applications_machine_learning
+    extra_args:
+      - machine_learning_SHIELD=pca63566
+      - FILE_SUFFIX="singlecore"
+      - CONFIG_PM=y
+      - CONFIG_PM_S2RAM=y
+      - CONFIG_PM_S2RAM_CUSTOM_MARKING=y

doc/nrf/releases_and_maturity/releases/release-notes-changelog.rst

@@ -360,7 +360,7 @@ nRF Desktop
 nRF Machine Learning (Edge Impulse)
 -----------------------------------
 
-|no_changes_yet_note|
+* Added power-optimized configuration for the :zephyr:board:`nrf54h20dk` board target.
 
 Serial LTE modem
 ----------------