[RFC] Implementing global run

[sternakt]

Why are these changes needed?

Related issue number

Checks

• I've included any doc changes needed for https://docs.ag2.ai/. See https://docs.ag2.ai/docs/contributor-guide/documentation to build and test documentation locally.
• I've added tests (if relevant) corresponding to the changes introduced in this PR.
• I've made sure all auto checks have passed.

[codecov]

Codecov Report

Attention: Patch coverage is 61.21372% with 147 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
autogen/run.py	46.42%	45 Missing ⚠️
autogen/agentchat/groupchat/groupchat.py	68.51%	12 Missing and 5 partials ⚠️
autogen/chat_managers/swarm.py	40.74%	14 Missing and 2 partials ⚠️
autogen/chat_managers/round_robin.py	38.09%	11 Missing and 2 partials ⚠️
autogen/agentchat/agent.py	47.05%	4 Missing and 5 partials ⚠️
autogen/messages/run_events.py	86.15%	9 Missing ⚠️
autogen/io/run_response.py	63.63%	0 Missing and 8 partials ⚠️
autogen/chat_managers/chat_manager.py	28.57%	2 Missing and 3 partials ⚠️
autogen/agentchat/conversable_agent.py	80.00%	2 Missing and 1 partial ⚠️
...ents/experimental/document_agent/document_agent.py	25.00%	3 Missing ⚠️
... and 10 more

❗ There is a different number of reports uploaded between BASE (338f89b) and HEAD (d78d37b). Click for more details.

HEAD has 1064 uploads less than BASE

Flag	BASE (338f89b)	HEAD (d78d37b)
3.9	71	0
ubuntu-latest	108	1
optional-deps	141	0
commsagent-slack	9	0
commsagent-discord	9	0
3.10	89	0
commsagent-telegram	9	0
3.13	75	0
core-without-llm	14	1
macos-latest	95	0
3.11	40	1
browser-use	3	0
3.12	28	0
windows-latest	100	0
jupyter-executor	9	0
graph-rag-falkor-db	6	0
twilio	9	0
retrievechat-pgvector	10	0
retrievechat	15	0
retrievechat-mongodb	10	0
retrievechat-qdrant	14	0
interop-crewai	9	0
docs	6	0
interop-pydantic-ai	9	0
websockets	9	0
crawl4ai	9	0
interop	9	0
interop-langchain	9	0
ollama	12	0
gpt-assistant-agent	3	0
teachable	3	0
agent-eval	1	0
gemini	11	0
llama-index-agent	3	0
websurfer	14	0
bedrock	10	0
cerebras	11	0
long-context	3	0
retrievechat-couchbase	3	0
anthropic	13	0
lmm	3	0
mistral	11	0
cohere	11	0
groq	14	0
together	9	0
swarm	8	0

Files with missing lines	Coverage Δ
autogen/__init__.py	100.00% <100.00%> (ø)
autogen/agentchat/__init__.py	100.00% <100.00%> (ø)
autogen/agentchat/groupchat/__init__.py	100.00% <100.00%> (ø)
.../agentchat/realtime/experimental/realtime_swarm.py	27.36% <100.00%> (ø)
autogen/agents/contrib/time/time_reply_agent.py	100.00% <100.00%> (ø)
...n/agents/experimental/reasoning/reasoning_agent.py	37.70% <100.00%> (ø)
autogen/chat_managers/__init__.py	100.00% <100.00%> (ø)
autogen/messages/agent_messages.py	96.37% <100.00%> (-0.48%)	⬇️
autogen/agentchat/termination.py	88.88% <88.88%> (ø)
autogen/agentchat/utils.py	80.44% <50.00%> (-1.02%)	⬇️
... and 18 more

... and 39 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[marklysze]

@sternakt / @davorrunje, can we:

• combine GroupRunPattern (Auto-speaker selection GroupChat) and SwarmRunPattern into the one GroupRunPattern and they both use initiate_swarm_chat within
• Can we support the other RunPatterns with the initiate_swarm_chat within. This is so that all RunPatterns have the same features, such as support for hand offs and shared context?

[marklysze]

Also - what's the current resume capability? E.g. if a run is stopped in the middle of a nested chat will it resume from there? Referring to nested chats as this and not just another run.

[davorrunje]

Closing this PR as the main part of it was already merged