Improvements to the EMAC test

[multiplemonomials]

Summary of changes

This PR does some refactoring and rewriting to improve the EMAC test. I have been working with this test suite a lot the last month or two and ended up making a number of fixes. This includes:

• Adding C structs that model the CTP packet format, so we can use those to read & write packets instead of just juggling all the pointers manually
• Making several of the tests slightly shorter and less verbose to make the test quicker to run
• Improving the multicast test to test removing addresses from the multicast filter
• Making some variable names and prints easier to understand

Impact of changes

Migration actions required

Documentation

---

Pull request type

[] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[] Feature update (New feature / Functionality change / New API)
[] Major update (Breaking change E.g. Return code change / API behaviour change)

---

Test results

[] No Tests required for this change (E.g docs only update)
[] Covered by existing mbed-os tests (Greentea or Unittest)
[] Tests / results supplied as part of this PR

---

[multiplemonomials]

50000 iterations made this test very long, and frankly I don't really see the point because there are only 1500 possible lengths that a packet can be, and that's more than enough to fuzz all of them.

[multiplemonomials]

I changed this so it will always print that a frame was received regardless of whether its format was correct, for ease of debugging

[multiplemonomials]

Ran the new EMAC test on K64F and NUCLEO_F429ZI. It works good! It correctly detected broken multicast filtering on both boards. The STM32F4 issue will be fixed in my next PR, but the K64F one is a longer term project as it's simply unimplemented.