fix(types): coerce dict server_tool_use to ServerToolUse in Usage init

Prerequisite for #31035 on this line, and a latent-bug fix in its own right.
#31035's usage-only fallback builds server_tool_use as a dict and prices it via
AnthropicConfig.calculate_usage, whose Usage(**model_dump()) round-trip drops it
back to a plain dict; without this Usage.__init__ coercion the recovered-cost path
does attribute access on a dict and raises, so #31035's web-search/server-tool
cost recovery is dead on arrival here. The same round-trip already affected the
pre-existing ChunkProcessor.calculate_usage path: every production consumer on
this line (litellm/llms/anthropic/cost_calculation.py,
litellm/litellm_core_utils/llm_cost_calc/tool_call_cost_tracking.py) reads
usage.server_tool_use.web_search_requests by attribute, so a dict there is a
latent AttributeError on streaming web-search cost. The coercion makes the value
a ServerToolUse, which all consumers expect.

Also updates the one test that pinned the old dict-subscript shape
(test_stream_chunk_builder_anthropic_web_search) to assert the ServerToolUse type
and attribute access, matching staging.

Content-verified present on litellm_internal_staging via aggregator
f49707bc66 (fix(otel) #30257), which carries both the coercion and the test
assertion update; this restores only those, not the rest of that aggregator. The
coercion also shipped to stable/1.89.x as 24e30b551f.

(cherry picked from commit 24e30b551f)

litellm/types/utils.py

@@ -1655,6 +1655,9 @@ class Usage(SafeAttributeModel, CompletionUsage):
             prompt_tokens_details=_prompt_tokens_details or None,
         )
 
+        if isinstance(server_tool_use, dict):
+            server_tool_use = ServerToolUse(**server_tool_use)
+
         if server_tool_use is not None:
             self.server_tool_use = server_tool_use
         else:  # maintain openai compatibility in usage object if possible

tests/test_litellm/litellm_core_utils/test_streaming_chunk_builder_utils.py

@@ -520,7 +520,10 @@ def test_stream_chunk_builder_anthropic_web_search():
     assert usage.prompt_tokens == 50
     assert usage.completion_tokens == 27
     assert usage.total_tokens == 77
-    assert usage.server_tool_use["web_search_requests"] == 2
+    # server_tool_use must be a ServerToolUse pydantic so downstream cost-calc
+    # (which uses attribute access) works. See issue #26153.
+    assert isinstance(usage.server_tool_use, ServerToolUse)
+    assert usage.server_tool_use.web_search_requests == 2
 
 
 def test_sort_chunks_handles_dict_hidden_params_created_at():