release: v0.6.0 — security audit fixes + 7 new features

Audit corrections (security, concurrency, stability):
- chat_engine: bound resp.Choices[0] access, release tool slot per-iteration
- conversation_multi: synchronous save under existing lock (was racy fire-and-forget)
- workflow/engine: short-circuit on failed deps (no more infinite busy-wait); track failed/skipped status
- handlers_workflow: rune-aware truncate for plan goal (UTF-8 safe)
- server: CORS limited to localhost origins (was wildcard)
- handlers_info / terminal: mask API keys and SSH passwords as "***" in GET responses; preserve stored secret if "***" sent on update
- terminal: sshpass uses -e + SSHPASS env var (was both -p and -e)
- handlers_chat: MaxBytesReader 50 MB on /api/chat
- image_cache: 10 MB cap per image
- handlers_config: font size <= 72; profile-save unmarshal errors propagated
- handlers_info: /lsp/auto-install ProjectDir restricted to user home
- Shell.jsx: parenthesized resize-condition (operator precedence)
- orchestrator_test: CleanAIResponse capitalization (fixes failing vet)

New features:
- platform: detect OS name (Debian, Ubuntu, Windows 11, macOS X.Y) and inject in Studio system prompt next to the date
- agents: default timeout 30 min for crush_run/claude_run (cap also 30 min)
- agents: new cwd, wsl_distro, wsl_user params; on Windows hosts launch via "wsl -d <distro> -u <user> --cd <cwd> --"
- agents: new claude_run tool (mirror of crush_run for Claude Code CLI)
- terminal: list installed WSL distros individually in new-tab menu (Windows only)
- studio: system prompt rewritten around BMAD-METHOD personas + mandatory delegation template
- studio: "Réflexion avancée" toggle — inactive provider produces a preliminary report injected as [RAPPORT PRÉALABLE] context for the active provider
- studio: "Historique compressé" toggle — collapses past tool calls to last action only, with "Tout afficher" expansion

CHANGELOG.md

@@ -4,6 +4,41 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
+## v0.6.0
+
+### Audit & corrections (sécurité, concurrence, stabilité)
+
+- fix(api): empty `resp.Choices[0]` panic in chat engine — bounded check
+- fix(api): `defer release()` accumulating inside tool-call loop — release immediately after each tool call
+- fix(api): race in `ConversationStoreMulti.Add` (fire-and-forget save under released lock) — synchronous save under existing lock
+- fix(workflow): infinite busy-wait in `engine.Execute` when a dependency fails — propagate `StatusFailed`/`StatusSkipped` and short-circuit
+- fix(workflow): UTF-8-unsafe slicing of plan goal — rune-aware truncate
+- fix(security): CORS `Access-Control-Allow-Origin: *` — restricted to localhost origins
+- fix(security): API key disclosure in `/api/providers` — masked as `"***"`; saving handler ignores `"***"` placeholder
+- fix(security): SSH password disclosure in `/api/terminal/sessions` — masked; update handler preserves stored password if `"***"` is sent
+- fix(security): sshpass `-p` + `-e` mutually-exclusive flags — use only `-e` with `SSHPASS` env var
+- fix(security): unbounded chat request body — `MaxBytesReader` 50 MB
+- fix(security): unbounded image upload — 10 MB cap in `saveImage`
+- fix(security): font size unbounded — capped at 72
+- fix(security): `LSP /auto-install` accepted arbitrary `project_dir` — restricted to user home subtree
+- fix(api): silent `json.Unmarshal` errors in profile save — propagated
+- fix(ui): operator-precedence bug in `Shell.jsx` resize check — parenthesized
+
+### Nouvelles fonctionnalités
+
+- feat(ai): inject OS name (e.g. `Debian 12`, `Windows 11`, `macOS 14.5`) alongside date in Studio system prompt
+- feat(agents): default timeout raised to 30 minutes for `crush_run` and `claude_run`; max also 30 min
+- feat(agents): new optional params `cwd`, `wsl_distro`, `wsl_user` — agents can be launched in a specific directory, and on Windows hosts inside a specific WSL distribution under a specific user
+- feat(agents): new `claude_run` tool (mirrors `crush_run` for the Claude Code CLI)
+- feat(terminal): WSL distros listed individually as quick-launch entries in the new-tab menu (Windows hosts only)
+- feat(studio): system prompt rewritten around the BMAD-METHOD (Analyst/PM/Architect/SM/Dev/QA personas + mandatory `[OBJECTIF]/[CONTEXTE]/[CONTRAINTES]/[LIVRABLE]/[CRITÈRE D'ACCEPTATION]` template for any agent delegation)
+- feat(studio): "Réflexion avancée" toggle — when enabled, the inactive AI provider produces a preliminary report that is injected as `[RAPPORT PRÉALABLE]` context into the active provider's prompt
+- feat(studio): "Historique compressé" toggle — collapses past tool calls and keeps only the last visible action per assistant message, with `Tout afficher` to expand
+
+### Bug fix CI
+
+- fix(test): `cleanAIResponse` → `CleanAIResponse` in `orchestrator_test.go` (was failing `go vet`)
+
 ## v0.4.0
 
 ### Changes since v0.3.5