[test gap] dualtor sending icmp requests with dstMAC==00:00:00:00:00:00

[zjswhhh]

Description

Some facts I noticed:

1. linkmgrd sends ICMP heartbeats with dstMAC==00:00:00:00:00:00 when server IP can't be resolved in active-standby scenarios.
2. FPGA icmp offload simply flips dst and src MAC.

This introduces a potential mac move risk, as when 2 servers under same rack are down together, if FPGA is still working, ToR will receive packet with srcMAC==00:00:00:00:00:00 from two ports.

Luckily that, I verified on HwSKU: Arista-7050CX3-32S-C32, zero mac wouldn't be learnt. It must be considered as invalid or a special MAC. To ensure this behavior doesn't change and avoid regression in future, we need a test case to cover the scenario above.

Steps to reproduce the issue:

Describe the results you received:

Describe the results you expected:

Additional information you deem important:

**Output of `show version`:**

```
(paste your output here)
```

**Attach debug file `sudo generate_dump`:**

```
(paste your output here)
```

[yxieca]

This should be a feature gap? SONiC could use different default macs for each port in such scenario?

[zjswhhh]

This should be a feature gap? SONiC could use different default macs for each port in such scenario?

Hi Ying - what default MAC should we use in active-standby scenario? We don't have MAC reserved for active-standby.