core: standalone signed-empty reasoning turns are replayed into later requests

[bbartels]

Description

V2 replays completed assistant turns whose only content is empty reasoning plus provider metadata.
In packages/core/src/session/runner/to-llm-message.ts, same-model reasoning retains provider metadata, and the meaningful filter considers metadata sufficient even when part.text === "". The existing omits empty assistant turns test explicitly preserves a standalone { text: "", state: { signature } } turn.
The Amazon Bedrock adapter serializes each retained part as:
{
"reasoningContent": {
"reasoningText": {
"text": "",
"signature": "..."
}
}
}
Controlled forks isolated the effect:

• The exact prompt with the clean full history (~620k provider tokens) produced visible text and tools.
• The same prompt with one or two standalone signed-empty turns still worked.
• The same prompt with five accumulated signed-empty turns deterministically returned another signature-only end_turn with no visible output.
• Manual compaction repaired the session because compaction rendering drops empty reasoning parts.
  Expected: signed reasoning state is retained when it accompanies semantic assistant output—text, readable reasoning, or a tool call—but a standalone metadata-only assistant turn is omitted from future model history or marked failed by the no-output handling described separately.
  Actual: state-only turns accumulate and can poison subsequent same-model requests.

Plugins

No response

OpenCode version

opencode2 v0.0.0-dev-18900; current V2 source af332a2 remains affected.

Steps to reproduce

1. Create a V2 session using a provider that returns signed reasoning metadata.
2. Persist several completed assistant turns containing only reasoning.text = "" and a provider signature.
3. Submit a prompt that works in a fork before those turns.
4. Observe the provider request replay each empty reasoningText plus signature.
5. Observe the later response repeat the signature-only pattern.
6. Compact the same session and retry; the prompt works because the empty reasoning turns are no longer replayed.

Screenshot and/or share link

No response

Operating System

No response

Terminal

No response

[github-actions]

This issue might be related to existing issues about empty or incomplete assistant turns being replayed to providers. Please check:

• Incomplete reasoning from a truncated turn is replayed to the provider #40147: Incomplete reasoning from a truncated turn is replayed to the provider
• Empty reasoning step (no delta) in replayed history causes 400 on later Kimi K3 requests #37651: Empty reasoning step (no delta) in replayed history causes 400 on later Kimi K3 requests
• Aborted assistant turn persisted with zero content bricks the session: empty assistant message replayed to provider (400 "must not be empty") #37946: Aborted assistant turn persisted with zero content bricks the session: empty assistant message replayed to provider
• Kimi (openai-compatible) rejects replayed reasoning+tool assistant turns: "message at position N with role 'assistant' must not be empty" #46577: Kimi rejects replayed reasoning+tool assistant turns: "message at position N with role 'assistant' must not be empty"

Your issue focuses specifically on signed empty reasoning turns (Bedrock provider metadata/signature accumulating across turns), which appears to be a distinct variant of the broader empty-assistant-turn replay problem. Linking these in case the fix should be coordinated.