(fix) get_response_headers for Azure OpenAI (#6344)

* fix get_response_headers

* unit testing for get headers

* unit testing for anthropic / azure openai headers

* increase test coverage for test_completion_response_ratelimit_headers

* fix test rate limit headers

tests/llm_translation/test_azure_openai.py

@@ -0,0 +1,96 @@
+import sys
+import os
+
+sys.path.insert(
+    0, os.path.abspath("../../")
+)  # Adds the parent directory to the system path
+
+import pytest
+from litellm.llms.AzureOpenAI.common_utils import process_azure_headers
+from httpx import Headers
+
+
+def test_process_azure_headers_empty():
+    result = process_azure_headers({})
+    assert result == {}, "Expected empty dictionary for no input"
+
+
+def test_process_azure_headers_with_all_headers():
+    input_headers = Headers(
+        {
+            "x-ratelimit-limit-requests": "100",
+            "x-ratelimit-remaining-requests": "90",
+            "x-ratelimit-limit-tokens": "10000",
+            "x-ratelimit-remaining-tokens": "9000",
+            "other-header": "value",
+        }
+    )
+
+    expected_output = {
+        "x-ratelimit-limit-requests": "100",
+        "x-ratelimit-remaining-requests": "90",
+        "x-ratelimit-limit-tokens": "10000",
+        "x-ratelimit-remaining-tokens": "9000",
+        "llm_provider-x-ratelimit-limit-requests": "100",
+        "llm_provider-x-ratelimit-remaining-requests": "90",
+        "llm_provider-x-ratelimit-limit-tokens": "10000",
+        "llm_provider-x-ratelimit-remaining-tokens": "9000",
+        "llm_provider-other-header": "value",
+    }
+
+    result = process_azure_headers(input_headers)
+    assert result == expected_output, "Unexpected output for all Azure headers"
+
+
+def test_process_azure_headers_with_partial_headers():
+    input_headers = Headers(
+        {
+            "x-ratelimit-limit-requests": "100",
+            "x-ratelimit-remaining-tokens": "9000",
+            "other-header": "value",
+        }
+    )
+
+    expected_output = {
+        "x-ratelimit-limit-requests": "100",
+        "x-ratelimit-remaining-tokens": "9000",
+        "llm_provider-x-ratelimit-limit-requests": "100",
+        "llm_provider-x-ratelimit-remaining-tokens": "9000",
+        "llm_provider-other-header": "value",
+    }
+
+    result = process_azure_headers(input_headers)
+    assert result == expected_output, "Unexpected output for partial Azure headers"
+
+
+def test_process_azure_headers_with_no_matching_headers():
+    input_headers = Headers(
+        {"unrelated-header-1": "value1", "unrelated-header-2": "value2"}
+    )
+
+    expected_output = {
+        "llm_provider-unrelated-header-1": "value1",
+        "llm_provider-unrelated-header-2": "value2",
+    }
+
+    result = process_azure_headers(input_headers)
+    assert result == expected_output, "Unexpected output for non-matching headers"
+
+
+def test_process_azure_headers_with_dict_input():
+    input_headers = {
+        "x-ratelimit-limit-requests": "100",
+        "x-ratelimit-remaining-requests": "90",
+        "other-header": "value",
+    }
+
+    expected_output = {
+        "x-ratelimit-limit-requests": "100",
+        "x-ratelimit-remaining-requests": "90",
+        "llm_provider-x-ratelimit-limit-requests": "100",
+        "llm_provider-x-ratelimit-remaining-requests": "90",
+        "llm_provider-other-header": "value",
+    }
+
+    result = process_azure_headers(input_headers)
+    assert result == expected_output, "Unexpected output for dict input"