Add library_name and update pipeline_tag

Hi! I'm Niels from the Hugging Face community team.

This PR improves the model card for SPARK by:
- Adding `library_name: transformers` to the metadata, as the model uses the Qwen2 architecture and is compatible with the Transformers library.
- Updating the `pipeline_tag` to `text-generation` to ensure the correct inference widget is displayed and to improve discoverability within the language model category.

These changes help users better interact with your model on the Hub.

README.md

@@ -1,12 +1,13 @@
 ---
-license: mit
+base_model:
+- Qwen/Qwen2.5-1.5B-Instruct
 language:
 - en
+license: mit
 metrics:
 - accuracy
-base_model:
-- Qwen/Qwen2.5-1.5B-Instruct
-pipeline_tag: reinforcement-learning
+library_name: transformers
+pipeline_tag: text-generation
 tags:
 - agent
 - reinforcement-learning