introduce CT_BITWISE_AND as CT_ARITH - #4510

Unfortunately this seems to not fix the issue in #4502 . The line in question is still formatted as if ((flags& MyEnum::FlagA) != MyEnum::Null) .
The parsing dump output also shows that the & is still parsed as a byref inside a c_cast, so it doesn't seem to be an issue with the config. Perhaps due to flags being parsed as TYPE even when it isn't and then we hit the assignment at line 1337 in these changes so the & is still set to CT_ADDR.

Is it worth adding the example given in #4502 as a test case? I can write it up properly as an additional test if needed.

I look...
Some more test data is always welcome!
Thanks

I look... Some more test data is always welcome! Thanks

I'll add a test and open it as a PR into this one

Edit: couldn't base it on this PR because it's coming from a fork too, but test at #4513

Any chance t get this fixes? Or is there a deeper problem?