Changelog

0.7.0

Added full service support for automations and scripts:
- proxmox_pve.start
- proxmox_pve.shutdown
- proxmox_pve.stop_hard
- proxmox_pve.reboot
Services are now fully multi-host capable:
- Automatic detection of the correct Proxmox host when using device_id
- Optional selection via config_entry_id
- Optional selection via host
- Automatic lookup by node/vmid/type if no host is given
- Clear error handling for ambiguous multi-host targets
Services can be used in:
- Automations
- Scripts
- Dashboards
Added services.yaml for proper UI descriptions in Home Assistant
Easy Proxmox can now be fully controlled without any buttons or switches, purely via automations

Extended Node monitoring with additional sensors:
- RAM Total (MB) and RAM Free (MB)
- Swap Used/Total/Free (MB)
- Node Storage (RootFS) Used/Total/Free in GB (3 decimals)
Kept existing Node sensors: CPU (%), Load (1m), RAM Used (MB), Uptime (d/h/m)
No changes to VM/CT entities, cleanup logic, or controls
Add README.md and LICENSE

Options are now applied live without restart or integration reload
- Changing polling interval updates all coordinators immediately
- Changing IP preference mode/prefix updates all existing guest coordinators immediately
Triggered refresh after saving options so sensors update quickly
Renamed integration to "Easy Proxmox (by René Bachmann)"

Fixed Options Flow crash that caused: “Config flow could not be loaded: 500 Internal Server Error”
Fixed incompatibility with Home Assistant’s OptionsFlow:
- Removed illegal assignment to the read-only config_entry property
- Now fetching the ConfigEntry safely via self.context["entry_id"]
Restored Options (gear icon) in the integration UI
Options dialog can now be opened and saved without backend errors
Improved compatibility with newer Home Assistant core versions
Stabilized Config Flow import and initialization

Added Options Flow:
- Configurable polling interval
- Configurable IP preference mode (prefer 192.168.*, private IPs, any, or custom prefix)
Added Proxmox Node devices:
- One device per Proxmox node
- Sensors for:
  - CPU usage (%)
  - RAM used (MB) and total RAM (attribute)
  - Uptime (days, hours, minutes)
  - Load average (1 minute)
VM/CT devices are now linked to their node device (via_device)
Existing dynamic VM handling, rename detection and hard cleanup retained

Fixed entity and device cleanup when a VM/CT is deleted:
- Entities are fully removed from Entity Registry
- Devices are fully removed from Device Registry
- No more “unavailable ghost entities”
Guaranteed hard cleanup for removed guests

Dynamic VM/CT discovery:
- New guests appear automatically without reload
- Removed guests are automatically cleaned up
Live rename handling:
- Device names and entity names update when VM name changes
Improved coordinator lifecycle handling