feat(teams): add progressive response lifecycle

[NeoHsu]

Important

Stacked draft: logical base stack/teams-06-processing-indicator is PR #1498. GitHub requires an upstream PR base to exist in openabdev/openab, so this draft temporarily targets main and may show preceding stack layers. Do not merge it until #1498 is merged and this branch is rebased onto current main; then review only its single incremental commit.

What problem does this solve?

Deliver progressive Teams answers without blind retries, duplicate finals, or destructive cleanup after an ambiguous write.

Discord Discussion URL: https://discord.com/channels/1491295327620169908/1491365158868619404/1531339032527765655
Microsoft Teams roadmap discussion.

Review Contract

Goal

Deliver progressive Teams answers without blind retries, duplicate finals, or destructive cleanup after an ambiguous write.

Non-goals

Long-message budgeting, attachment ingress, durable activity ownership, and automatic recovery queues are deferred.

Accepted Residual Risks

A placeholder can remain after Unknown; a rejected edit can require one fresh final send only when the prior outcome is authoritative; no operation can prove cross-device presentation.

Acceptance Criteria

The feature is explicit and default off; placeholder creation returns a real ID; same-content writes deduplicate; overflow and cleanup are ordered; Unknown suppresses later writes; the ACK-loss harness cannot activate without narrow guards.

Follow-ups

Add exact Teams message budgeting in PR 9 and consider durable reconciliation independently.

At a Glance

Authenticated Teams activity / operator target
                    │
                    ▼
            [this PR’s bounded layer]
                    │
                    ▼
          explicit outcome or fail-closed stop

Prior Art & Industry Research

OpenClaw: its Microsoft Teams extension separates access checks, Bot Framework route context, and outbound operations. For this slice the relevant comparison is progressive message updates and ambiguous-write handling.

Hermes Agent: its Teams platform adapter keeps Teams-specific transport and message shaping behind a platform adapter. It does not provide OpenAB’s negotiated Core/Gateway outcome contract, so this PR keeps the useful adapter boundary but adds explicit fail-closed semantics.

Proposed Solution

• Add default-off progressive Teams responses using one owned placeholder.
• Make every transition outcome-aware and stop on ambiguity.
• Add a guarded ACK-loss proxy harness for deterministic failure-path testing.

Why this approach?

A state machine over explicit outcomes makes each write and cleanup decision auditable, including ACK-loss cases.

Alternatives Considered

Always fresh-send after an edit problem (rejected: duplicates after Unknown) or delete first (rejected: can destroy the only delivered content).

Validation

• cargo check -p openab-core
• cargo check -p openab-gateway --features teams
• python3 scripts/test-teams-ack-drop-proxy.py (12 tests at this stack layer; 14 after PR 9)