Fix external module crash for att open proxy scanner

[adfoster-r7]

Fix external module crash for when running the auxiliary/scanner/wproxy/att_open_proxy module

Verification

Before

The att_open_proxy module iterated over an RHOST string, instead of an array of RHOST ips

msf6 auxiliary(scanner/wproxy/att_open_proxy) > run 127.0.0.1

[-] .:49152 - Error connecting: encoding with 'idna' codec failed (UnicodeError: label empty or too long)
[-] .:49152 - Error connecting: encoding with 'idna' codec failed (UnicodeError: label empty or too long)
[-] 1:49152 - Error connecting: [Errno 65] No route to host
[-] 7:49152 - Error connecting: [Errno 65] No route to host
[-] 2:49152 - Error connecting: [Errno 65] No route to host
[-] .:49152 - Error connecting: encoding with 'idna' codec failed (UnicodeError: label empty or too long)
[-] 1:49152 - Error connecting: [Errno 65] No route to host
[-] [:49152 - Error connecting: [Errno 8] nodename nor servname provided, or not known
[-] ":49152 - Error connecting: [Errno 8] nodename nor servname provided, or not known
[-] ]:49152 - Error connecting: [Errno 8] nodename nor servname provided, or not known
[-] 0:49152 - Error connecting: [Errno 61] Connect call failed ('0.0.0.0', 49152)
[-] 0:49152 - Error connecting: [Errno 61] Connect call failed ('0.0.0.0', 49152)
[-] ":49152 - Error connecting: [Errno 8] nodename nor servname provided, or not known
[*] Scanned 1 of 1 hosts (100% complete)
[*] Auxiliary module execution completed

After

Verifying the at&t open proxy scanner now works:

# Create a fake listener in a tab
print -n "\x2a\xce..." | ncat -lnvp 49152 

# Run the module - no longer crashes, and you can see the ncat client has connected
msf6 auxiliary(scanner/wproxy/att_open_proxy) > run

[+] 127.0.0.1:49152 - Matches
[*] Scanned 1 of 1 hosts (100% complete)
[*] Auxiliary module execution completed

Also verifying get user spns still works - which was originally updated to work in #17490

msf6 auxiliary(gather/get_user_spns) > run rhost=192.168.123.13 domain=adf3.local user=Administrator PASS=p4$$w0rd4

[*] Running for 192.168.123.13...
[+] ServicePrincipalName                    Name            MemberOf  PasswordLastSet             LastLogon                   Delegation    
[+] --------------------------------------  --------------  --------  --------------------------  --------------------------  -------------
[+] fake_msql/dc3.adf3.local:1433           fake_mysql                2022-02-10 10:57:13.356981  <never>                     unconstrained 
[+] fake_msql/dc3.adf3.local                fake_mysql                2022-02-10 10:57:13.356981  <never>                     unconstrained 
[+] adf3.local/kerberoastable_service       kerberoastable            2022-06-27 20:03:33.970786  2022-06-30 14:49:29.019696                
[+] adf3.local/kerberoastable_service:1337  kerberoastable            2022-06-27 20:03:33.970786  2022-06-30 14:49:29.019696                
[+] $krb5tgs$23$*fake_mysql$ADF3.LOCAL$adf3.local/fake_mysql*$8ecdb0028c2afe5fe22....

[adfoster-r7]

I originally added this in #17490 - but I believe this PR is a cleaner solution

The real issue is the external python modules use a datastore that doesn't have the same semantics as the Ruby version; i.e. i.e. RHOST/RHOSTS aliasing support, plus case insensitive lookups

[adfoster-r7]

Required because the original datastore had a delete method that also removed data options (Which had other bugs associated with it)

https://github.com/rapid7/metasploit-framework/pull/16940/files#diff-09925245579a8207d4933e78a8e76c5168fe2a1134f0b02a13f9ff344e2c8dc4

[dwelch-r7]

Release Notes

Fix external module crash for when running the auxiliary/scanner/wproxy/att_open_proxy module