feat: add Event.intermediate flag for multi-step agent streaming

[nuthalapativarun]

Please ensure you have read the contribution guide before creating a pull request.

Link to Issue or Description of Change

• Related: Streaming SSE yields intermediate LLM text during multi-tool-call turns — no way to distinguish intermediate vs final text events #261

Problem:
During multi-step agent turns (e.g. model text → tool call → model text), streaming consumers cannot distinguish model text emitted before a tool call from the final response text. Both appear as identical events, forcing clients to buffer everything or show intermediate reasoning text to the user unintentionally.

Solution:
Add an optional intermediate?: boolean field to the Event interface. In LlmAgent.runAsyncImpl, events from steps that are followed by function calls are marked intermediate: true. Final response events remain unmarked (undefined).

for await (const event of runner.runAsync({ ... })) {
  if (event.intermediate) continue; // skip reasoning steps
  displayToUser(event);
}

This is a purely additive, non-breaking change. Existing consumers that do not check intermediate are unaffected. Aligned with adk-python behaviour.

Implementation details:

• Event.intermediate?: boolean added to core/src/events/event.ts
• LlmAgent.runAsyncImpl in core/src/agents/llm_agent.ts now collects all events per step, checks whether the step is final (isFinalResponse), and sets intermediate: true on non-final steps before yielding

Testing Plan

Unit Tests:

• I have added or updated unit tests for my change.
• All unit tests pass locally.

Summary of passed npm test results:

✓ |unit:core| core/test/agents/event_intermediate_test.ts (5 tests) 7ms

Test Files  1 passed (1)
     Tests  5 passed (5)

Full suite (on this branch): 1101 passed | 21 skipped

Manual End-to-End (E2E) Tests:
The intermediate field is optional and additive. All existing tests pass without modification.

Checklist

• I have read the CONTRIBUTING.md document.
• I have performed a self-review of my own code.
• I have commented my code, particularly in hard-to-understand areas.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• I have manually tested my changes end-to-end.
• Any dependent changes have been merged and published in downstream modules.

[nuthalapativarun]

Hi team, gentle ping on this one — happy to address any feedback or make adjustments if needed. Thanks for reviewing!

[nuthalapativarun]

Hi team, gentle ping on this one — happy to address any feedback or make adjustments if needed. Thanks for reviewing!