docs(tools): document automatic tool output truncation

[webwww123]

What does this PR do?

Fixes #4826.

Adds a dedicated section to tools docs explaining automatic tool-output truncation:

• large outputs are truncated before being fed back into context
• full output is persisted under ~/.local/share/opencode/tool-output/
• agents are guided to use grep/read with offset/limit on the saved file

Also clarifies there is no stable user-facing threshold tuning config today.

How did you verify your code works?

• Cross-checked behavior against current truncation implementation and tool output handling in the codebase.
• Validation will run in GitHub Actions CI.

[github-actions]

The following comment was made by an LLM, it may be inaccurate:

Based on my search, I found related PRs about tool output handling and truncation, but no direct duplicates of PR #13119 (the current PR):

Related PRs (not duplicates):

• feat(opencode): switch tool truncation to token-based limits #9648 - feat(opencode): switch tool truncation to token-based limits - Related to the underlying truncation implementation being documented
• fix(opencode): preserve task metadata after truncation #11905 - fix(opencode): preserve task metadata after truncation - Related to truncation behavior
• fix(session): prevent context overflow by adding safety margin to compaction check #6562 - fix(session): prevent context overflow by adding safety margin to compaction check - Related to context overflow prevention

These are implementation-focused PRs that address the technical aspects of tool output truncation, whereas PR #13119 is specifically about documenting this feature. They are complementary but not duplicates.

No duplicate PRs found

[github-actions]

Closing this pull request because it has had no updates for more than 60 days. If you plan to continue working on it, feel free to reopen or open a new PR.