allow waiting for all function calls to complete

[aconchillo]

Please describe the changes in your PR. If it is addressing an issue, please reference that as well.

This PR adds a wait_for_all flag to the LLMService base class. This flag allows waiting for all function calls to complete before returning any result to the LLM. So, when wait_for_all is True we won't return any IN_PROGRESS result to the LLM.

This is necessary for reasoning models as both the thinking and tool result are part of the same assistant turn.

[codecov[bot]]

Codecov Report

:x: Patch coverage is 17.64706% with 28 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/pipecat/services/llm_service.py	17.64%	28 Missing :warning:

Files with missing lines	Coverage Δ
src/pipecat/services/llm_service.py	37.20% <17.64%> (-1.88%)	:arrow_down:

:rocket: New features to boost your workflow:

• :snowflake: Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[aconchillo]

Did some more tests and fixed function call cancellation issues.