Update release notes with new features and enhancements (#612)

- Adds details to the RL interfaces.
- Adds new export occupancy grid system.

Signed-off-by: Arjo Chakravarty <[email protected]>

Author: arjo129

jetty/release_notes.md

@@ -64,8 +64,8 @@ the implementation pull request
 ## Improve Gazebo APIs for Reinforcement Learning
 
 Several APIs have been improved to streamline reinforcement learning pipelines.
-We have also added an example integration with StableBaselines3 for
-reinforcement learning, enabling users to experiment with RL algorithms within
+We have also added an [example integration with StableBaselines3](https://github.com/gazebosim/gz-sim/tree/gz-sim10/examples/scripts/reinforcement_learning/simple_cart_pole)
+for reinforcement learning, enabling users to experiment with RL algorithms within
 Gazebo. The example provides a starting point for developing and testing robotic
 control policies.
 
@@ -107,3 +107,10 @@ See the [tracking issue](https://github.com/gazebosim/gz-sim/issues/1260) and
 one of the implementation pull requests
 <https://github.com/gazebosim/gz-sim/pull/2992> for more details
 
+## Occupancy Grid Export
+
+This enhancement adds a plugin that enables end users to directly export occupancy
+grids for use with Nav2 or other mobile robotics software. 
+
+See the pull request <https://github.com/gazebosim/gz-sim/pull/2958> for more details.
+