feat(#2305): hive-screen-mcp — screenshot + input MCP for GUI agents

New crate hive-screen-mcp: a stdio MCP bridge activated automatically
when an agent has hyperhive.gui.enable = true. Provides five tools:

- screenshot   — grim → saves PNG, returns path for Read tool
- type_text    — wtype → Unicode text input (no daemon)
- key_press    — ydotool key → combos like ctrl+c, super+l
- mouse_move   — ydotool mousemove --absolute
- mouse_click  — ydotool click, optionally with prior move

New nix/agent-modules/screen.nix: wires the MCP bridge into
extraMcpServers.screen; adds grim + wtype to systemPackages. Adds
hyperhive.gui.screenInput option (default false) which enables the
ydotoold daemon + ydotool for mouse/keyboard injection via /dev/uinput.

screenshot and type_text work without screenInput. key_press,
mouse_move, and mouse_click return a ydotool error until ydotoold is
running and /dev/uinput is accessible in the container.

hive-screen-mcp/Cargo.toml

@@ -0,0 +1,23 @@
+[package]
+name = "hive-screen-mcp"
+edition.workspace = true
+version.workspace = true
+
+[lints]
+workspace = true
+
+[dependencies]
+anyhow.workspace = true
+rmcp.workspace = true
+schemars.workspace = true
+serde.workspace = true
+tokio.workspace = true
+tracing.workspace = true
+tracing-subscriber.workspace = true
+
+# `hive-screen-mcp` — stdio MCP bridge for GUI agents.
+# Provides screenshot (grim), text input (wtype), key press and mouse
+# actions (ydotool) for agents running a Weston compositor.
+[[bin]]
+name = "hive-screen-mcp"
+path = "src/main.rs"