fix(components): normalize ConditionAgent response content before parsing

[jslim-23]

Summary

ConditionAgent currently assumes response.content is always a string when parsing the model output. In LangChain v1, response.content may also be returned as (ContentBlock | ContentBlock.Text)[], which can cause JSON parsing to fail in ConditionAgent.

This change normalizes the response content before parsing by reusing the existing extractResponseContent() helper, keeping response handling consistent with other parts of the codebase.

The parse error path is also updated to report the same normalized content instead of casting response.content to string again in the catch block.

Changes

Use extractResponseContent(response) before calling parseJsonMarkdown() in packages/components/nodes/agentflow/ConditionAgent/ConditionAgent.ts

Reuse the normalized response content in analytics, parsing, output content, and parse error reporting

How to reproduce the bug

1. Create an Agentflow with a ConditionAgent node
2. Use a LangChain v1-compatible chat model that returns response.content as text content blocks
3. Run the flow
4. ConditionAgent may fail while parsing the model output as JSON

[gemini-code-assist]

Code Review

This pull request refactors the ConditionAgent to extract the LLM response content into a single variable, responseContent, which is then reused for analytics, JSON parsing, and the final output. This change improves consistency and reduces redundant processing. A suggestion was provided to truncate the raw response content in error logs to enhance readability when parsing fails.

[gemini-code-assist]

For better readability in error logs, it's a good practice to truncate long raw responses. The responseContent could be quite large, making the error message difficult to read. Consider truncating it if it exceeds a certain length, similar to how it's done in parseJsonBody.

Suggested change

	`Failed to parse a valid scenario from the LLM's response. Please check if the model is capable of following JSON output instructions. Raw LLM Response: "${responseContent}"`
	"Failed to parse a valid scenario from the LLM's response. Please check if the model is capable of following JSON output instructions. Raw LLM Response: \"" + responseContent.substring(0, 200) + (responseContent.length > 200 ? "..." : "") + "\""

References

1. Prioritize code readability and understandability over conciseness. Truncating large blobs in logs improves the readability and utility of error messages for debugging.