Add `--release` tests to our CI

[mhasel]

We should add release testing to our CI to get ahead of future issues regarding release builds.

[mhasel]

Currently blocked by

• [x] Tests causing segfaults:

SIGSEGV [   0.099s] rusty::tests correctness::functions::mux_string_literal
SIGSEGV [   0.091s] rusty::tests correctness::functions::mux_string_ref
SIGSEGV [   0.060s] rusty::tests correctness::functions::sel_string_literal
SIGSEGV [   0.058s] rusty::tests correctness::functions::sel_string_ref
SIGSEGV [   0.046s] rusty::tests correctness::strings::string_returned_from_generic_wrapper_function_does_not_truncate
SIGSEGV [   0.125s] iec61131std::string_conversion_tests string_to_wstring_conversion
SIGSEGV [   0.123s] iec61131std::string_conversion_tests string_to_wstring_extra_conversion
SIGSEGV [   0.112s] iec61131std::string_conversion_tests wstring_to_string_conversion
SIGSEGV [   0.110s] iec61131std::string_conversion_tests wstring_to_string_extra_conversion
SIGSEGV [   0.110s] iec61131std::string_function_tests delete_ext_string_with_escape_sequence
SIGSEGV [   0.087s] iec61131std::string_function_tests delete_string_with_escape_sequence
SIGSEGV [   0.094s] iec61131std::string_function_tests left_wstring

tracked in #1121

• [x] Bitaccess returning incorrect values for booleans

FAIL [   0.035s] iec61131std::bit_conversion_tests byte_to_bool
FAIL [   0.035s] iec61131std::bit_conversion_tests dword_to_bool
FAIL [   0.028s] iec61131std::bit_conversion_tests lword_to_bool
FAIL [   0.034s] iec61131std::bit_conversion_tests word_to_bool

tracked in #1141

• [ ] date_time_numeric_functions::checked_mul_time_with_signed_int returning garbage values

FAIL [   0.025s] iec61131std::date_time_numeric_functions_tests mul_ltime_signed
FAIL [   0.032s] iec61131std::date_time_numeric_functions_tests mul_signed
FAIL [   0.028s] iec61131std::date_time_numeric_functions_tests mul_time_signed

tracked in #1146

• [ ] result of validation_functions::is_valid_bcd failing expected assertion (assert!(maintype.valid);)

FAIL [   0.029s] iec61131std::validation_functions_tests is_valid_word

• [ ] date_time_extra_functions::concat_dt panicking due to hitting .expect("Invalid parameters, cannot create TOD") in function body (in1 shows garbage values in the debugger)

FAIL [   0.045s] iec61131std::date_time_extra_functions_tests concat_dt_signed_ints
FAIL [   0.039s] iec61131std::date_time_extra_functions_tests concat_dt_unsigned_ints
FAIL [   0.056s] iec61131std::date_time_extra_functions_tests concat_ldt_signed_ints
FAIL [   0.042s] iec61131std::date_time_extra_functions_tests concat_ldt_unsigned_ints

iec61131std::validation_functions_tests is_valid_word passes if a println!() is added on the conditional expression. For iec61131std::date_time_extra_functions_tests concat_dt_signed_ints (and most likely the other 3, but I haven't tried this for all of them) the tests can be "fixed" by bitwise-anding arguments to date_time_extra_functions::concat_tod with 0xFFFF, so the problem is most likely the upcast from i16 to u32 in date_time_extra_functions::CONCAT_TOD__INT. A simple println!() on the upcasted parameter in2 of MUL_LTIME__SINT also results in the correct value being reported. This is reproducible also when directly compiling an ST example with our stdlib compiled with release.