Update pipeline tag to text-generation

This PR updates the `pipeline_tag` in the model card metadata from `text-classification` to `text-generation`.

The MedVAL-4B model takes task instructions, original input, and AI-generated output, and then *generates* structured textual feedback including `reasoning`, specific `errors` (with explanations), and a `risk_level`. This output format, which involves generating explanatory text rather than a single classification label, aligns more accurately with the `text-generation` pipeline. This change will improve the discoverability and accurate categorization of the model on the Hugging Face Hub.

README.md

@@ -11,9 +11,9 @@ license: mit
 metrics:
 - f1
 - accuracy
+pipeline_tag: text-generation
 tags:
 - medical
-pipeline_tag: text-classification
 ---
 
 **MedVAL-4B** (medical text validator) is a language model fine-tuned to **assess AI-generated medical text** outputs at near **physician-level reliability**.