Add community evaluation results for SWE-BENCH_PRO, TERMINAL-BENCH-2.0

This PR adds community-provided evaluation results for the following benchmarks:

- **[SWE-BENCH_PRO](https://huggingface.co/datasets/ScaleAI/SWE-bench_Pro)**
- **[TERMINAL-BENCH-2.0](https://huggingface.co/datasets/harborframework/terminal-bench-2.0)**

These results were extracted from the model card. This is based on the new [evaluation results feature](https://huggingface.co/docs/hub/eval-results).

*Note: This is an automated PR. Please review the evaluation results before merging.*

.eval_results/swe-bench_pro.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: ScaleAI/SWE-bench_Pro
+    task_id: SWE_Bench_Pro
+  value: 56.2
+  source:
+    url: https://huggingface.co/MiniMaxAI/MiniMax-M2.7
+    name: Model Card

.eval_results/terminal-bench-2.0.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: harborframework/terminal-bench-2.0
+    task_id: terminalbench_2
+  value: 57
+  source:
+    url: https://huggingface.co/MiniMaxAI/MiniMax-M2.7
+    name: Model Card