Allow to specify different network_ids for AvalancheGo and the Ash CLI

[ngurmen]

I completed our setup using the ash ansible collection on fuji. I used the following parameters and the whole setup works really well.

...
# Avalanche context
ash_cli_avalanche_network_id: fuji
...
# network selection
avalanchego_network_id: fuji
...

However, transferring this setup to our production network gave a lot of problems. Just using mainnet instead of fuji in the above parameters gives the following error:

RPC error: failed to get Subnets for network 'mainnet': unknown RPC error

And it fails at the Get the list of blockchains tracked by the node (1/2) task.

If I change the parameters as below it gives the same error above.

...
# Avalanche context
ash_cli_avalanche_network_id: mainnet-ankr
...
# network selection
avalanchego_network_id: mainnet
...

If I change both ids to mainnet-ankr it assumes that it is a custom network because of the lines below

ansible-avalanche-collection/roles/node/tasks/config-node.yml

Line 9 in 1bae7d6

when: avalanchego_network_id not in ['mainnet', 'testnet', 'fuji']

ansible-avalanche-collection/roles/node/tasks/config-node.yml

Line 20 in 1bae7d6

} if avalanchego_network_id not in ['mainnet', 'testnet', 'fuji'] else {} }}

and starts running that branch but fails with the fallowing error

'''dict object'' has no attribute ''bootstrap_nodes'''

If I use mainnet-ankr for both ids the ash.avalanche.health_checks playbook works fine but ash.avalanche.provision_nodes playbook gives the bootstrap_nodes error above.

If I add 'mainnet-ankr' to the list of networks on the above L9 and L20 lines seem to work but actually the node does not start because the node.json created ends up getting that network name which avalanchego does not recognize.

Could you please suggest a way forward? Is this a configuration issue on my part?