fix(models): pass NOT_GIVEN instead of None for system instruction in AnthropicLlm

When no system instruction is set (e.g. during event compaction),
system_instruction is None. The Anthropic API rejects None — it
expects a str or list of content blocks. This causes a 400 Bad
Request when compaction fires via LlmEventSummarizer.

Pass NOT_GIVEN (already imported) when system_instruction is
None/empty so the parameter is omitted from the API call. Fixes
both streaming and non-streaming code paths.

Fixes #5318

Author: anmolg1997

src/google/adk/models/anthropic_llm.py

@@ -402,10 +402,16 @@ async def generate_content_async(
         else NOT_GIVEN
     )
 
+    system = (
+        llm_request.config.system_instruction
+        if llm_request.config and llm_request.config.system_instruction
+        else NOT_GIVEN
+    )
+
     if not stream:
       message = await self._anthropic_client.messages.create(
           model=model_to_use,
-          system=llm_request.config.system_instruction,
+          system=system,
           messages=messages,
           tools=tools,
           tool_choice=tool_choice,
@@ -431,9 +437,14 @@ async def _generate_content_streaming(
     a final aggregated LlmResponse with all content.
     """
     model_to_use = self._resolve_model_name(llm_request.model)
+    system = (
+        llm_request.config.system_instruction
+        if llm_request.config and llm_request.config.system_instruction
+        else NOT_GIVEN
+    )
     raw_stream = await self._anthropic_client.messages.create(
         model=model_to_use,
-        system=llm_request.config.system_instruction,
+        system=system,
         messages=messages,
         tools=tools,
         tool_choice=tool_choice,