Improve execute_tool errors for mismatched tool input key casing

[saif-at-scalekit]

Problem

Developers can confuse Gmail response field names with tool input parameter names. For example, gmail_fetch_mails response objects include camelCase fields such as threadId, but gmail_get_thread_by_id expects the tool input key thread_id.

If a caller passes {"threadId": "abc123"}, the current error reports that input.thread_id is missing. That is technically accurate, but it sends developers toward checking whether the value was extracted instead of noticing that the key casing is wrong.

Example

# Response extraction uses camelCase
thread_id = message.get("threadId", "")

# Tool input must use the schema key
response = client.actions.execute_tool(
    tool_name="gmail_get_thread_by_id",
    connected_account_id=connected_account_id,
    tool_input={"thread_id": thread_id},
)

Suggested fix

When execute_tool receives a required-field error and the submitted input contains a likely case-converted variant, return a more actionable message, for example:

Required field 'thread_id' not found. Did you mean 'threadId'? Tool input keys must match the snake_case schema shown for this tool.

If the Python SDK does not have enough schema context at this layer, this should be coordinated with the API/service error response and surfaced cleanly by the SDK.

Developer impact

This reduces a common debug loop when chaining tool responses into follow-up tool calls, especially for Gmail fields such as threadId -> thread_id and messageId -> message_id.