[Snappi] Fixes for T2 ixia test cases: Traffic MAC address, peer_ports, SnappiFanout

[sdszhang]

Description of PR

Summary: Fixes :

1. Chose gateway MAC based on port, as the gateway MAC maybe different on packet chassis.
2. change peer_ports calculation in get_multidut_tgen_peer_port_set to cover all the dut/asic combination.
3. SnappiFanoutManager only supports Snappi chassis, so ignore other type of fanout switches during initialization.
4. change the default prefix length to /31 (ipv4), /126(ipv6). as these are the default prefix length that will be advertised in SONiC BGP for connected routes.

Type of change

• [ ] Bug fix
• [ ] Testbed and Framework(new/improvement)
• [ ] Test case(new/improvement)

Back port request

• [ ] 202012
• [ ] 202205
• [ ] 202305
• [ ] 202311
• [x] 202405

Approach

What is the motivation for this PR?

Fixes needed to run test_multidut_pfcwd_a2a_with_snappi.py

How did you do it?

1. Chose gateway MAC based on port, as the gateway MAC maybe different on packet chassis.
2. change peer_ports calculation in get_multidut_tgen_peer_port_set to cover all the dut/asic combination.
3. SnappiFanoutManager only supports Snappi chassis, so ignore other type of fanout switches during initialization.

How did you verify/test it?

Run test manually.

Any platform specific information?

Supported testbed topology if it's a new test case?

Documentation

[mssonicbld]

Cherry-pick PR to 202405: https://github.com/sonic-net/sonic-mgmt/pull/13570