feat: add Discussion mode and refine workspace Explorer

[elkaix]

Related Issue

Related to #251.

Depends on #263 for the workspace Explorer foundation used by the Discussion UI.

Problem

Pythinker needs one bounded two-model discussion flow that behaves the same in Web and the interactive TUI. Users also need to see where the models agree or diverge and open workspace files without mixing file controls with task controls.

What changed

• Add experimental Discussion mode with persistent session-local Architect and Builder configuration.
• Run independent opinions concurrently, then allow optional Architect comparison and optional fresh Architect Fusion.
• Show comparison results as green Agreement, amber Divergence, and blue Final analysis cards.
• Keep a paused Discussion on a static mascot with Waiting… until the user starts the next stage.
• Replace the task controls with a dedicated files sidebar after Show files, with Back to tasks, workspace search, lazy browsing, and file preview.
• Show a consistent hover and keyboard-focus label for every icon-only sidebar action.
• Keep /discussion canonical and /expert-opinion compatible across Web, Desktop, and the interactive TUI.

Enable with PYTHINKER_CODE_EXPERIMENTAL_EXPERT_TALK=1, then run /discussion.

Budget decision

The staged flow keeps 8 read-only tool rounds per opening and 4 per optional comparison or Fusion stage. This gives an explicit maximum of 56 provider attempts and supersedes the earlier 24-attempt planning cap in #251.

Verification

• Web: 95 files and 1,147 tests passed.
• Focused agent-core-v2 regression: 12 tests passed.
• Focused TUI panel regression: 4 tests passed.
• Root typecheck, Web typecheck, and lint passed with 0 errors.
• Web build, committed bundle parity, and VitePress documentation build passed.
• Electron packaging and strict Developer ID signature verification passed.
• Live Electron verification covered sidebar icon labels, the Show files click path, files search and preview, Back to tasks, comparison cards, and the paused Waiting… state.

Checklist

• I have read the CONTRIBUTING document.
• I have linked a related issue (external PRs: the issue must have a maintainer's /approve).
• I have added tests that prove my feature works.
• Ran gen-changesets skill, or this PR needs no changeset.
• Ran gen-docs skill, or this PR needs no doc update.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

🗂️ Base branches to auto review (1)

• main

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: b65a9b5c-1e4d-4688-9ca0-f3619e8140d0

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

_{Comment @coderabbitai help to get the list of available commands.}

[pkg-pr-new]

pnpm dlx https://pkg.pr.new/@pymodel/pythinker-code@3b74d77

npx https://pkg.pr.new/@pymodel/pythinker-code@3b74d77

commit: 3b74d77