core/tools: show images instead of describing them

New `view_image` reads a local picture (JPEG, PNG, GIF, WebP, sniffed from its
bytes) and hands it to the model as an image part, with an optional `detail`
level — `low` measured 3.4x cheaper on a 1200x900 picture. `read_file` now
points at it instead of returning decode noise.

`fetch_local` returns an image response as the picture itself, and MCP image
content blocks become image parts instead of `[image: ...]` placeholders, so a
screenshot server finally reaches the model.

tests/test_cli/test_mcp_cmd.py

@@ -68,7 +68,7 @@ async def test_slash_mcp_list_statuses(state: ReplState, capsys: pytest.CaptureF
         out = capsys.readouterr().out
         assert "fake" in out
         assert "connected" in out
-        assert "tools=3" in out
+        assert "tools=4" in out
     finally:
         await manager.aclose()
 
@@ -92,7 +92,7 @@ async def test_slash_mcp_reconnect_connects_new_config(
         statuses = state.mcp_manager.statuses()
         assert [name for name, _, _ in statuses] == ["fake"]
         assert statuses[0][1] == "connected"
-        assert statuses[0][2] == 3
+        assert statuses[0][2] == 4
         out = capsys.readouterr().out
         assert "tools=off" in out
         assert "fake" in out