[Docs] Add AMD GPU info to ssh fleets section

Issue: #1413

docs/docs/concepts/fleets.md

@@ -91,9 +91,14 @@ are both acceptable).
 
     !!! info "Requirements" 
         Hosts should be pre-installed with Docker.
+        The user should have passwordless `sudo` access.
+
         Systems with NVIDIA GPUs should also be pre-installed with CUDA 12.1 and
         [NVIDIA Container Toolkit :material-arrow-top-right-thin:{ .external }](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html).
-        The user should have `sudo` access.
+
+        Systems with AMD GPUs should also be pre-installed with AMDGPU-DKMS kernel driver via
+        [native package manager :material-arrow-top-right-thin:{ .external }](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/native-install/index.html)
+        or [AMDGPU installer :material-arrow-top-right-thin:{ .external }](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/amdgpu-install.html).
 
     ??? info "Environment variables"
         For SSH fleets, it's possible to pre-configure environment variables.