feat: add persisted manual pi backend mode controls

README.md

@@ -363,6 +363,15 @@ backends:
 
 `pi_embedded` is intended for canary migration cohorts. In spike mode (`no_tools_mode: true`), Flynn keeps tool-oriented turns on native and only routes plain-text turns to Pi.
 
+Runtime backend mode can be controlled live (persisted in `~/.local/share/flynn/preferences.json`):
+
+- `/backend status` shows runtime mode and effective backend selection
+- `/backend activate pi` forces `pi_embedded` globally
+- `/backend deactivate pi` forces native for Pi-routed turns
+- `/backend use config` resets to `backends.default`
+
+This manual runtime mode control is the intended Pi activation/deactivation switch.
+
 To evaluate canary performance from audit logs, run:
 
 ```bash
@@ -602,6 +611,7 @@ Notes:
 | `/approve [id]` | Approve latest (or specific) pending gate |
 | `/deny [id] [reason]` | Deny latest (or specific) pending gate |
 | `/skill <list|search|install>` | In-chat skill discovery/install (`list`, `search <term>`, `install <registry-id>`) |
+| `/backend <status\|activate pi\|deactivate pi\|use config>` | Show or control global runtime backend mode |
 
 ## Web UI Dashboard
 
@@ -637,7 +647,7 @@ pnpm tui:fs
 | `/model` | Show all model tiers and which is active |
 | `/model <tier>` | Switch active tier (`local`, `default`, `fast`, `complex`, or aliases `ollama`, `sonnet`, `haiku`, `opus`) |
 | `/model <tier> <provider/model>` | Hot-swap a tier's provider and model at runtime |
-| `/backend [provider]` | Show or switch local backend (`ollama`, `llamacpp`) |
+| `/backend [provider]` | TUI-local command: show or switch local model backend (`ollama`, `llamacpp`) |
 | `/login [provider]` | Authenticate with GitHub (OAuth device flow) |
 | `/reset` | Clear history |
 | `/status` | Show session info |