feat: enable streaming usage metrics for OpenAI-compatible providers (#4326)

Inject `stream_options={"include_usage": True} `when streaming and
OpenTelemetry telemetry is active. Telemetry always overrides any caller
preference to ensure complete and consistent observability metrics.

Changes:
- Add conditional stream_options injection to OpenAIMixin (benefits
OpenAI, Bedrock, Runpod, Together, Fireworks providers)
- Add conditional stream_options injection to LiteLLMOpenAIMixin
(benefits WatsonX and other litellm-based providers)
- Check telemetry status using trace.get_current_span().is_recording()
- Override include_usage=False when telemetry active to prevent metric
gaps
- Unit tests for this functionality

Fixes #3981

Note: this work originated in PR #4200, which I closed after rebasing on
the telemetry changes. This PR rebases those commits, incorporates the
Bedrock feedback, and carries forward the same scope described there.
## Test Plan
#### OpenAIMixin + telemetry injection tests 
PYTHONPATH=src python -m pytest
tests/unit/providers/utils/inference/test_openai_mixin.py

#### LiteLLM OpenAIMixin tests
PYTHONPATH=src python -m pytest
tests/unit/providers/inference/test_litellm_openai_mixin.py -v

#### Broader inference provider
PYTHONPATH=src python -m pytest tests/unit/providers/inference/
--ignore=tests/unit/providers/inference/test_inference_client_caching.py
-v

tests/unit/providers/utils/inference/test_openai_compat.py

@@ -0,0 +1,68 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from unittest.mock import MagicMock, patch
+
+from llama_stack.providers.utils.inference.openai_compat import (
+    get_stream_options_for_telemetry,
+)
+
+
+class TestGetStreamOptionsForTelemetry:
+    def test_returns_original_when_not_streaming(self):
+        stream_options = {"keep": True}
+
+        result = get_stream_options_for_telemetry(stream_options, False)
+
+        assert result is stream_options
+
+    def test_streaming_without_active_span_returns_original(self):
+        stream_options = {"keep": True}
+
+        with patch("opentelemetry.trace.get_current_span", return_value=None):
+            result = get_stream_options_for_telemetry(stream_options, True)
+
+        assert result is stream_options
+
+    def test_streaming_with_inactive_span_returns_original(self):
+        stream_options = {"keep": True}
+        mock_span = MagicMock()
+        mock_span.is_recording.return_value = False
+
+        with patch("opentelemetry.trace.get_current_span", return_value=mock_span):
+            result = get_stream_options_for_telemetry(stream_options, True)
+
+        assert result is stream_options
+
+    def test_streaming_with_active_span_adds_usage_when_missing(self):
+        mock_span = MagicMock()
+        mock_span.is_recording.return_value = True
+
+        with patch("opentelemetry.trace.get_current_span", return_value=mock_span):
+            result = get_stream_options_for_telemetry(None, True)
+
+        assert result == {"include_usage": True}
+
+    def test_streaming_with_active_span_merges_existing_options(self):
+        stream_options = {"other_option": "value"}
+        mock_span = MagicMock()
+        mock_span.is_recording.return_value = True
+
+        with patch("opentelemetry.trace.get_current_span", return_value=mock_span):
+            result = get_stream_options_for_telemetry(stream_options, True)
+
+        assert result == {"other_option": "value", "include_usage": True}
+        assert stream_options == {"other_option": "value"}
+
+    def test_streaming_with_active_span_overrides_include_usage_false(self):
+        stream_options = {"include_usage": False}
+        mock_span = MagicMock()
+        mock_span.is_recording.return_value = True
+
+        with patch("opentelemetry.trace.get_current_span", return_value=mock_span):
+            result = get_stream_options_for_telemetry(stream_options, True)
+
+        assert result["include_usage"] is True