Description of problem:

OCP docs:
https://docs.openshift.com/container-platform/4.17/networking/cidr-range-definitions.html

ROSA docs:
https://docs.openshift.com/rosa/networking/cidr-range-definitions.html#machine-cidr-description

Current default subnets are:

join switch subnet :
    V4JoinSubnet:       "100.64.0.0/16",
     V6JoinSubnet:       "fd98::/64",
Transit switch subnet:
    V4TransitSwitchSubnet: "100.88.0.0/16",
    V6TransitSwitchSubnet: "fd97::/64",
Masquerade subnet :
   defaultV4MasqueradeSubnet = "169.254.0.0/17"
   defaultV6MasqueradeSubnet = "fd69::/112

Currently, in the docs it is stated:
For Red Hat OpenShift Service on AWS 4.17 and later versions, clusters use 169.254.0.0/17 for IPv4 and fd69::/112 for IPv6 as the default masquerade subnet. These ranges should also be avoided by users. For upgraded clusters, there is no change to the default masquerade subnet.
The default subnets and are customizable at day 2 from Openshift 4.14.38 onward.

This pertains to all 3: Join, Transit and Masquerade subnets

Update the documentation to mention why we need those subnets.

Masquerade subnet is used to avoid same source and destination IP when the traffic originates from a node and gets hair-pinned to the same node after the routing decision is made.

join switch is required to connect a distributed logical router to a non distributed logical router.