fix: handle server-ip-within-range case correctly

[starbops]

IMPORTANT: Please do not create a Pull Request without creating an issue first.

Problem:

Given an IPPool object with the following YAML:

apiVersion: network.harvesterhci.io/v1alpha1
kind: IPPool
metadata:
  namespace: default
  name: test-net
spec:
  ipv4Config:
    cidr: 192.168.30.0/24
  networkName: default/test-net

The mutator will fill the missing .spec.ipv4Config.serverIP field, and since there's no router specified, the serverIP will be 192.168.30.1 by default. However, that IP address is within the allocatable range (as the mutator-filled .spec.ipv4Config.pool.start and .spec.ipv4Config.pool.end fields will be 192.168.30.1 and 192.168.30.254, respectively.) In the previous version, though the server IP address will be marked as RESERVED under the .status.ipv4.allocated field, it wasn't handled adequately by the webhook and controller:

• The validating admission webhook treats the RESERVED IP address the same way as allocated IP addresses. It will fail when validating any further updating requests because it thinks the server IP address is already occupied.
• The cache-building control loop does not skip the RESERVED IP addresses. They should be skipped just like how we did to the excluded IPs.

Solution:

If the server IP is within the allocatable pool range, it will be marked as "RESERVED" and honored by the webhook and controller's cache-builder reconciliation loop.

• Splitting the allocated, excluded, and reserved IP address lists
• The validating webhook only checks if the server IP address falls into the allocated and excluded IP address lists
• The cache-building control loop skips to allocate reserved IP addresses

Related Issue:

harvester/harvester#5225

Test plan:

1. Install and enable the harvester-vm-dhcp-controller add-on

   apiVersion: harvesterhci.io/v1beta1
   kind: Addon
   metadata:
     labels:
       addon.harvesterhci.io/experimental: "true"
     namespace: harvester-system
     name: harvester-vm-dhcp-controller
   spec:
     chart: harvester-vm-dhcp-controller
     enabled: true
     repo: https://charts.harvesterhci.io
     valuesContent: |
       image:
         repository: starbops/harvester-vm-dhcp-controller
         tag: fix-5225-head
       agent:
         image:
           repository: starbops/harvester-vm-dhcp-agent
           tag: fix-5225-head
       webhook:
         image:
           repository: starbops/harvester-vm-dhcp-webhook
           tag: fix-5225-head
     version: 0.3.0

2. Prepare a VM Network (NAD) named test-net
3. Create a minimal IPPool object associated to the VM Network

   apiVersion: network.harvesterhci.io/v1alpha1
   kind: IPPool
   metadata:
     namespace: default
     name: test-net
   spec:
     ipv4Config:
       cidr: 192.168.30.0/24
     networkName: default/test-net

4. Check the auto assigned DHCP server IP address. It should be recorded in .spec.ipv4Config.serverIP and marked as RESERVED under .status.ipv4.allocated
5. Create a new VM and attach to the test-net VM Network
6. A new IP address should be allocated for the VM and recorded under .status.ipv4.allocated field
7. Delete the VM
8. Delete the IPPool object
9. Check if the agent Pod harvester-system/default-test-net-agent is gone

[w13915984028]

LGTM, thanks.