feat: [new open benchmark] DocVQA (#1647)

# What does this PR do?
DocVQA asks model to look a a picture, then answer a question given in
text, with a text answer by text information in the picture. these
questions often require understanding of relative positions of texts
within the picture.

original dataset is defined in the "Task1" of
https://www.docvqa.org/datasets


## Test Plan
setup llama server with 

```
llama stack run ./llama_stack/templates/open-benchmark/run.yaml
```


then send traffic:

```
 llama-stack-client eval run-benchmark "meta-reference-docvqa"  --model-id   meta-llama/Llama-3.3-70B-Instruct     --output-dir /tmp/gpqa    --num-examples   200
```

llama_stack/templates/open-benchmark/open_benchmark.py

@@ -203,6 +203,13 @@ def get_distribution_template() -> DistributionTemplate:
                 uri="huggingface://datasets/llamastack/bfcl_v3?split=train",
             ),
         ),
+        DatasetInput(
+            dataset_id="docvqa",
+            purpose=DatasetPurpose.eval_messages_answer,
+            source=URIDataSource(
+                uri="huggingface://datasets/llamastack/docvqa?split=val",
+            ),
+        ),
     ]
 
     default_benchmarks = [
@@ -231,6 +238,11 @@ def get_distribution_template() -> DistributionTemplate:
             dataset_id="bfcl",
             scoring_functions=["basic::bfcl"],
         ),
+        BenchmarkInput(
+            benchmark_id="meta-reference-docvqa",
+            dataset_id="docvqa",
+            scoring_functions=["basic::docvqa"],
+        ),
     ]
     return DistributionTemplate(
         name=name,