feat: [new open benchmark] BFCL_v3 (#1578)

# What does this PR do?
create a new dataset BFCL_v3 from
https://gorilla.cs.berkeley.edu/blogs/13_bfcl_v3_multi_turn.html

overall each question asks the model to perform a task described in
natural language, and additionally a set of available functions and
their schema are given for the model to choose from. the model is
required to write the function call form including function name and
parameters , to achieve the stated purpose. the results are validated
against provided ground truth, to make sure that the generated function
call and the ground truth function call are syntactically and
semantically equivalent, by checking their AST .



## Test Plan

start server by 

```
llama stack run ./llama_stack/templates/ollama/run.yaml
```

then send traffic
```
 llama-stack-client eval run-benchmark "bfcl"  --model-id   meta-llama/Llama-3.2-3B-Instruct    --output-dir /tmp/gpqa    --num-examples   2
```




[//]: # (## Documentation)

llama_stack/templates/open-benchmark/open_benchmark.py

@@ -226,6 +226,22 @@ def get_distribution_template() -> DistributionTemplate:
                 "chat_completion_input": {"type": "string"},
             },
         ),
+        DatasetInput(
+            dataset_id="bfcl",
+            provider_id="huggingface",
+            url=URL(uri="https://huggingface.co/datasets/llamastack/bfcl_v3"),
+            metadata={
+                "path": "llamastack/bfcl_v3",
+                "split": "train",
+            },
+            dataset_schema={
+                "function": {"type": "string"},
+                "language": {"type": "string"},
+                "ground_truth": {"type": "string"},
+                "id": {"type": "string"},
+                "chat_completion_input": {"type": "string"},
+            },
+        ),
     ]
 
     default_benchmarks = [
@@ -249,6 +265,11 @@ def get_distribution_template() -> DistributionTemplate:
             dataset_id="math_500",
             scoring_functions=["basic::regex_parser_math_response"],
         ),
+        BenchmarkInput(
+            benchmark_id="meta-reference-bfcl",
+            dataset_id="bfcl",
+            scoring_functions=["basic::bfcl"],
+        ),
     ]
     return DistributionTemplate(
         name=name,

llama_stack/templates/open-benchmark/run.yaml

@@ -216,6 +216,24 @@ datasets:
     split: test
   dataset_id: math_500
   provider_id: huggingface
+- dataset_schema:
+    function:
+      type: string
+    language:
+      type: string
+    ground_truth:
+      type: string
+    id:
+      type: string
+    chat_completion_input:
+      type: string
+  url:
+    uri: https://huggingface.co/datasets/llamastack/bfcl_v3
+  metadata:
+    path: llamastack/bfcl_v3
+    split: train
+  dataset_id: bfcl
+  provider_id: huggingface
 scoring_fns: []
 benchmarks:
 - dataset_id: simpleqa
@@ -238,6 +256,11 @@ benchmarks:
   - basic::regex_parser_math_response
   metadata: {}
   benchmark_id: meta-reference-math-500
+- dataset_id: bfcl
+  scoring_functions:
+  - basic::bfcl
+  metadata: {}
+  benchmark_id: meta-reference-bfcl
 tool_groups:
 - toolgroup_id: builtin::websearch
   provider_id: tavily-search