Bedrock + Anthropic extended thinking: error because thinking parts are not sent back

[DouweM]

From @snakingfire in #2040 (comment):

reasoning_model = BedrockConverseModel(
        "us.anthropic.claude-sonnet-4-20250514-v1:0",
        provider=BedrockProvider(bedrock_client=bedrock_client),
    )
reasoning_model_settings = BedrockModelSettings(
        parallel_tool_calls=True,
        bedrock_additional_model_requests_fields={
            "thinking": {"type": "enabled", "budget_tokens": 1024}
        },
    )
...

botocore.errorfactory.ValidationException: An error occurred (ValidationException) when calling the Converse operation: The model returned the following errors: messages.1.content.0.type: Expected thinking or redacted_thinking, but found text. When thinking is enabled, a final assistant message must start with a thinking block (preceeding the lastmost set of tool_use and tool_result blocks). We recommend you include thinking blocks from previous turns. To avoid this requirement, disable thinking. Please consult our documentation at https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking

Like #2072 but for Bedrock

[pydanticai-bot]

PydanticAI Github Bot Found 2 issues similar to this one:

1. "Anthropic extended thinking: invalid request error when using tools #2040" (90% similar)
2. "Unable to use Thinking AND tool calling with AnthropicModel #2075" (90% similar)