Fix a Listener Bug

[water111]

We've had a few random test failures related to the listener connection on github, and recently a test for format that sent a lot of data and failed always on github actions. It seems like the listener can send data faster than the runtime can read it, causing an error that I wasn't checking for. After adding the error check, the format test seems to pass.

If all tests pass on this PR, I think this is the fix.

[water111]

Lots of failures!

• Linux failure is due to Coveralls, the tests passed.
• Windows failed once because of listener reasons
• Windows failed a second time because format is totally broken there. I suspect I forgot to update some assembly stuff for windows.

[water111]

BEEF appears:

[DECI2] Got message:
 3886 0 0xe042 H -> E
[DECI2] Got message:
 10572 47870 0xbeef � -> �
[DECI2] Got message:
 24 0 0xe042 H -> E

The beef is a placeholder in instructions, so I think the Listener is sending code twice (unlikely?) or the message receive is aborting too early and receiving part way through a message.

[coveralls]

Pull Request Test Coverage Report for Build 254871209

• 2 of 2 (100.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 59.417%

---

Totals
Change from base Build 254788801:	0.0%
Covered Lines:	12439
Relevant Lines:	20935

---

💛 - Coveralls

[water111]

Looks like there are two separate bugs. This fixes a stupid mistake in the listener, the other bug is here: #46 and I'll fix it tomorrow.