(feat) Add usage tracking for streaming /anthropic passthrough routes (#6842)

* use 1 file for AnthropicPassthroughLoggingHandler

* add support for anthropic streaming usage tracking

* ci/cd run again

* fix - add real streaming for anthropic pass through

* remove unused function stream_response

* working anthropic streaming logging

* fix code quality

* fix use 1 file for vertex success handler

* use helper for _handle_logging_vertex_collected_chunks

* enforce vertex streaming to use sse for streaming

* test test_basic_vertex_ai_pass_through_streaming_with_spendlog

* fix type hints

* add comment

* fix linting

* add pass through logging unit testing

litellm/llms/anthropic/chat/handler.py

@@ -779,3 +779,32 @@ class ModelResponseIterator:
             raise StopAsyncIteration
         except ValueError as e:
             raise RuntimeError(f"Error parsing chunk: {e},\nReceived chunk: {chunk}")
+
+    def convert_str_chunk_to_generic_chunk(self, chunk: str) -> GenericStreamingChunk:
+        """
+        Convert a string chunk to a GenericStreamingChunk
+
+        Note: This is used for Anthropic pass through streaming logging
+
+        We can move __anext__, and __next__ to use this function since it's common logic.
+        Did not migrate them to minmize changes made in 1 PR.
+        """
+        str_line = chunk
+        if isinstance(chunk, bytes):  # Handle binary data
+            str_line = chunk.decode("utf-8")  # Convert bytes to string
+            index = str_line.find("data:")
+            if index != -1:
+                str_line = str_line[index:]
+
+        if str_line.startswith("data:"):
+            data_json = json.loads(str_line[5:])
+            return self.chunk_parser(chunk=data_json)
+        else:
+            return GenericStreamingChunk(
+                text="",
+                is_finished=False,
+                finish_reason="",
+                usage=None,
+                index=0,
+                tool_use=None,
+            )