fix(buzz-agent): Responses reasoning summary, Anthropic display:summarized, ACP v2 messageId

[wpfleger96]

Three pre-existing gaps in the buzz-agent observer feed fixed together per Will's ruling ("all 3 in the current PR"):

1. OpenAI/DBv2-GPT route — responses_body never requested reasoning.summary; GPT-family models billed thinking tokens but returned summary: [].
2. Anthropic/DBv2-Claude route — anthropic_thinking_config() never sent thinking.display; newest Claude models (Opus 5, Sonnet 5, Fable 5, Mythos 5, Opus 4.7/4.8, Mythos Preview) default to display:"omitted", returning thinking blocks with an empty thinking field — observer rendered nothing.
3. ACP v2 compliance — buzz-agent negotiates ACP v2 but emitted agent_thought_chunk and agent_message_chunk without messageId, which ACP v2's ContentChunk requires (messageId + content both required at schema head d13d1baa).

Changes

crates/buzz-agent/src/config.rs

• New ThinkingSummary enum (Auto/Concise/Detailed) with BUZZ_AGENT_THINKING_SUMMARY env var (default Auto); mirrors BUZZ_AGENT_THINKING_EFFORT pattern
• anthropic_thinking_config() now emits "display": "summarized" in both the adaptive shape and the manual-budget shape whenever thinking is enabled
• Rewrote is_adaptive_thinking_model and anthropic_thinking_config doc comments to match Anthropic's exact three-way per-model terminology (doc: https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#supported-models):
  • Opus 4.6/4.7/4.8, Sonnet 4.6: Off — thinking OFF by default; type:"adaptive" required to enable
  • Opus 5, Sonnet 5: On — thinking on by default, can be disabled; we still send type:"adaptive" to activate output_config.effort
  • Fable 5, Mythos 5, Mythos Preview: Always on — thinking cannot be disabled; we still send type:"adaptive" to activate output_config.effort

crates/buzz-agent/src/llm.rs

• responses_body emits reasoning.summary alongside reasoning.effort when effort is set (gated — no bare reasoning:{summary} without effort)
• Covers both the pure-OpenAI Responses path and the DBv2 GPT-family Responses path

crates/buzz-agent/src/agent.rs

• agent_thought_chunk carries "messageId": format!("{run_id}-thought-{round}")
• agent_message_chunk carries "messageId": format!("{run_id}-message-{round}")
• The two IDs are distinct (thought and assistant are two logical messages per the ACP v2 Message ID RFD)
• run_id is a fresh random token per session/prompt invocation so IDs are session-unique across multiple prompts

crates/buzz-agent/src/lib.rs

• run_id plumbed into RunCtx (was already generated in run_prompt, just not threaded through)

crates/buzz-agent/tests/golden_transcripts.rs

• test_acp_v2_chunks_carry_message_id — negotiates v2, drives two consecutive session/prompt calls, asserts: both chunk types carry non-empty messageId; thought and message IDs are distinct; IDs do not recur across the two prompts in the same ACP session

desktop/src-tauri/src/managed_agents/env_vars.rs

• BUZZ_AGENT_THINKING_SUMMARY added to is_safe_to_reveal allowlist

desktop/src-tauri/src/commands/agent_config_tests.rs

• Tests for BUZZ_AGENT_THINKING_SUMMARY allowlist entry (case-insensitive)

Tests added

• parse_thinking_summary_round_trips_all_values
• parse_thinking_summary_unset_and_empty_yield_auto
• parse_thinking_summary_is_case_insensitive
• parse_thinking_summary_rejects_unknown_value
• thinking_summary_as_str_mapping
• responses_body_summary_present_iff_effort_set
• responses_body_emits_configured_summary_mode
• responses_body_concise_summary_mode
• anthropic_thinking_config_adaptive_emits_display_summarized
• anthropic_thinking_config_manual_budget_emits_display_summarized
• test_acp_v2_chunks_carry_message_id (integration test — two-prompt cross-session case)

Notes

• DBv2 gateway parity for display: unverified — the DBv2 Claude route proxies Anthropic Messages shape, but whether the gateway passes thinking.display through is not confirmed. Flagged here rather than blocking on it.
• buzz-acp and Desktop TS are unchanged — they already parse messageId as optional and will pick it up from the wire automatically.
• Chat Completions and OpenRouter paths: untouched.

[wesbillman]

Reviewed on Wes's behalf at exact head 47ef696d6094d5fa4adcdc5a0a3dbe63de0a2727. No blocking findings.

I traced all three changes through request construction, provider-response parsing, and ACP emission:

• OpenAI Responses: reasoning.summary is emitted only alongside configured reasoning.effort; parse_responses already consumes every summary_text item. Default auto is the provider-recommended way to select the most detailed summary supported by the model. Explicit concise/detailed remain operator choices because OpenAI documents model-dependent support.
• Anthropic: display: "summarized" is valid in both type: "adaptive" and type: "enabled" shapes. Anthropic's current thinking documentation explicitly supports both modes and explains that newer-model defaults otherwise return an empty thinking field. The field remains absent when this code does not enable/configure thinking.
• ACP v2: each emitted thought and assistant message gets a distinct ID, stable across its chunks and unique across rounds and prompt invocations because the prefix is the existing fresh per-prompt run_id. The two-prompt golden test pins the collision boundary that the earlier implementation missed.

The new env value is non-secret and its reveal allowlist/config parsing are covered. git diff --check passed. Full relevant CI is green (Rust Lint, Unit Tests, Desktop Core/build/smoke, Mobile, Windows Rust, cross-compiles, Security); two Desktop integration shards were still running when reviewed. The author-disclosed DBv2 Anthropic gateway passthrough uncertainty may affect whether that route gains observer text, but it does not create a regression in the native Anthropic path and is not a merge blocker.

Verdict: 9/10, safe to merge once required CI completes. This is a comment-only review, not an approval.