feat(providers): sambanova safety provider (#2221)

# What does this PR do?

Includes SambaNova safety adaptor to use the sambanova cloud served
Meta-Llama-Guard-3-8B
minor updates in sambanova docs

## Test Plan
pytest -s -v tests/integration/safety/test_safety.py
--stack-config=sambanova --safety-shield=sambanova/Meta-Llama-Guard-3-8B

llama_stack/templates/sambanova/build.yaml

@@ -1,6 +1,6 @@
 version: '2'
 distribution_spec:
-  description: Use SambaNova for running LLM inference
+  description: Use SambaNova for running LLM inference and safety
   providers:
     inference:
     - remote::sambanova
@@ -10,7 +10,7 @@ distribution_spec:
     - remote::chromadb
     - remote::pgvector
     safety:
-    - inline::llama-guard
+    - remote::sambanova
     agents:
     - inline::meta-reference
     telemetry:

llama_stack/templates/sambanova/doc_template.md

@@ -37,33 +37,44 @@ The following models are available by default:
 
 ### Prerequisite: API Keys
 
-Make sure you have access to a SambaNova API Key. You can get one by visiting [SambaNova.ai](https://sambanova.ai/).
+Make sure you have access to a SambaNova API Key. You can get one by visiting [SambaNova.ai](http://cloud.sambanova.ai?utm_source=llamastack&utm_medium=external&utm_campaign=cloud_signup).
 
 
 ## Running Llama Stack with SambaNova
 
 You can do this via Conda (build code) or Docker which has a pre-built image.
 
-### Via Docker
 
-This method allows you to get started quickly without having to build the distribution code.
+### Via Docker
 
 ```bash
 LLAMA_STACK_PORT=8321
+llama stack build --template sambanova --image-type container
 docker run \
   -it \
-  --pull always \
   -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
-  llamastack/distribution-{{ name }} \
+  -v ~/.llama:/root/.llama \
+  distribution-{{ name }} \
   --port $LLAMA_STACK_PORT \
   --env SAMBANOVA_API_KEY=$SAMBANOVA_API_KEY
 ```
 
+
+### Via Venv
+
+```bash
+llama stack build --template sambanova --image-type venv
+llama stack run --image-type venv ~/.llama/distributions/sambanova/sambanova-run.yaml \
+  --port $LLAMA_STACK_PORT \
+  --env SAMBANOVA_API_KEY=$SAMBANOVA_API_KEY
+```
+
+
 ### Via Conda
 
 ```bash
 llama stack build --template sambanova --image-type conda
-llama stack run ./run.yaml \
+llama stack run --image-type conda ~/.llama/distributions/sambanova/sambanova-run.yaml \
   --port $LLAMA_STACK_PORT \
   --env SAMBANOVA_API_KEY=$SAMBANOVA_API_KEY
 ```

llama_stack/templates/sambanova/run.yaml

@@ -38,10 +38,11 @@ providers:
       user: ${env.PGVECTOR_USER:}
       password: ${env.PGVECTOR_PASSWORD:}
   safety:
-  - provider_id: llama-guard
-    provider_type: inline::llama-guard
+  - provider_id: sambanova
+    provider_type: remote::sambanova
     config:
-      excluded_categories: []
+      url: https://api.sambanova.ai/v1
+      api_key: ${env.SAMBANOVA_API_KEY}
   agents:
   - provider_id: meta-reference
     provider_type: inline::meta-reference
@@ -189,6 +190,9 @@ models:
   model_type: embedding
 shields:
 - shield_id: meta-llama/Llama-Guard-3-8B
+  provider_shield_id: sambanova/Meta-Llama-Guard-3-8B
+- shield_id: sambanova/Meta-Llama-Guard-3-8B
+  provider_shield_id: sambanova/Meta-Llama-Guard-3-8B
 vector_dbs: []
 datasets: []
 scoring_fns: []

llama_stack/templates/sambanova/sambanova.py

@@ -34,7 +34,7 @@ def get_distribution_template() -> DistributionTemplate:
     providers = {
         "inference": ["remote::sambanova", "inline::sentence-transformers"],
         "vector_io": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
-        "safety": ["inline::llama-guard"],
+        "safety": ["remote::sambanova"],
         "agents": ["inline::meta-reference"],
         "telemetry": ["inline::meta-reference"],
         "tool_runtime": [
@@ -110,7 +110,7 @@ def get_distribution_template() -> DistributionTemplate:
     return DistributionTemplate(
         name=name,
         distro_type="self_hosted",
-        description="Use SambaNova for running LLM inference",
+        description="Use SambaNova for running LLM inference and safety",
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
@@ -122,7 +122,15 @@ def get_distribution_template() -> DistributionTemplate:
                     "vector_io": vector_io_providers,
                 },
                 default_models=default_models + [embedding_model],
-                default_shields=[ShieldInput(shield_id="meta-llama/Llama-Guard-3-8B")],
+                default_shields=[
+                    ShieldInput(
+                        shield_id="meta-llama/Llama-Guard-3-8B", provider_shield_id="sambanova/Meta-Llama-Guard-3-8B"
+                    ),
+                    ShieldInput(
+                        shield_id="sambanova/Meta-Llama-Guard-3-8B",
+                        provider_shield_id="sambanova/Meta-Llama-Guard-3-8B",
+                    ),
+                ],
                 default_tool_groups=default_tool_groups,
             ),
         },