feat: Support Gen AI attributes for Amazon Nova foundational model (#300)

*Description of changes:*
Added GenAI inference parameters auto instrumentation support for Amazon
Nova.

Contract tests:
<img width="924" alt="image"
src="https://github.com/user-attachments/assets/5e7544d7-3d37-432f-a1d7-25690f12453f"
/>

<img width="889" alt="image"
src="https://github.com/user-attachments/assets/408f4c7a-5d69-41d1-800a-b88f8db7aef0"
/>

By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice.

Author: liustve

aws-opentelemetry-distro/src/amazon/opentelemetry/distro/patches/_bedrock_patches.py

@@ -268,6 +268,8 @@ def extract_attributes(self, attributes: _AttributeMapT):
 
                     if "amazon.titan" in model_id:
                         self._extract_titan_attributes(attributes, request_body)
+                    if "amazon.nova" in model_id:
+                        self._extract_nova_attributes(attributes, request_body)
                     elif "anthropic.claude" in model_id:
                         self._extract_claude_attributes(attributes, request_body)
                     elif "meta.llama" in model_id:
@@ -288,6 +290,12 @@ def _extract_titan_attributes(self, attributes, request_body):
         self._set_if_not_none(attributes, GEN_AI_REQUEST_TOP_P, config.get("topP"))
         self._set_if_not_none(attributes, GEN_AI_REQUEST_MAX_TOKENS, config.get("maxTokenCount"))
 
+    def _extract_nova_attributes(self, attributes, request_body):
+        config = request_body.get("inferenceConfig", {})
+        self._set_if_not_none(attributes, GEN_AI_REQUEST_TEMPERATURE, config.get("temperature"))
+        self._set_if_not_none(attributes, GEN_AI_REQUEST_TOP_P, config.get("top_p"))
+        self._set_if_not_none(attributes, GEN_AI_REQUEST_MAX_TOKENS, config.get("max_new_tokens"))
+
     def _extract_claude_attributes(self, attributes, request_body):
         self._set_if_not_none(attributes, GEN_AI_REQUEST_MAX_TOKENS, request_body.get("max_tokens"))
         self._set_if_not_none(attributes, GEN_AI_REQUEST_TEMPERATURE, request_body.get("temperature"))
@@ -324,6 +332,7 @@ def _set_if_not_none(attributes, key, value):
         if value is not None:
             attributes[key] = value
 
+    # pylint: disable=too-many-branches
     def on_success(self, span: Span, result: Dict[str, Any]):
         model_id = self._call_context.params.get(_MODEL_ID)
 
@@ -342,6 +351,8 @@ def on_success(self, span: Span, result: Dict[str, Any]):
                 response_body = json.loads(telemetry_content.decode("utf-8"))
                 if "amazon.titan" in model_id:
                     self._handle_amazon_titan_response(span, response_body)
+                if "amazon.nova" in model_id:
+                    self._handle_amazon_nova_response(span, response_body)
                 elif "anthropic.claude" in model_id:
                     self._handle_anthropic_claude_response(span, response_body)
                 elif "meta.llama" in model_id:
@@ -375,6 +386,17 @@ def _handle_amazon_titan_response(self, span: Span, response_body: Dict[str, Any
                 if "completionReason" in result:
                     span.set_attribute(GEN_AI_RESPONSE_FINISH_REASONS, [result["completionReason"]])
 
+    # pylint: disable=no-self-use
+    def _handle_amazon_nova_response(self, span: Span, response_body: Dict[str, Any]):
+        if "usage" in response_body:
+            usage = response_body["usage"]
+            if "inputTokens" in usage:
+                span.set_attribute(GEN_AI_USAGE_INPUT_TOKENS, usage["inputTokens"])
+            if "outputTokens" in usage:
+                span.set_attribute(GEN_AI_USAGE_OUTPUT_TOKENS, usage["outputTokens"])
+        if "stopReason" in response_body:
+            span.set_attribute(GEN_AI_RESPONSE_FINISH_REASONS, [response_body["stopReason"]])
+
     # pylint: disable=no-self-use
     def _handle_anthropic_claude_response(self, span: Span, response_body: Dict[str, Any]):
         if "usage" in response_body: