FMSLogo Code

Fix "FORWARD 100 5*" to run FORWARD 100 before throwing a "not enough inputs" error.
This is part of fixing Bug #250.

This is not 100% correct, as the infix expressions prior to the error won't
be evaluated in something like "1*2*3+4+". However, I don't this can have a
visible impact, as the built-in infix operators don't have side effects and
redefining them turns them into prefix operator.

Fix a typo in a comment

Fix "FORWARD 100 )" to run FORWARD 100 before throwing a "unexpected ')'" error.
This is part of fixing Bug #250.

This has an unexpected consequence of making "FORWARD)" throw a "not enough inputs to FORWARD"
error instead of a "unexpected ')'" error, as it used to. I don't think this is bad because
both errors apply and technically FORWARD was missing an input before the ")" is reached.

This latter change is why EVALTEST.UNEXPECTEDCLOSEPAREN was modified to move the right paren.
However, I did test that the new definition still causes an infinite loop in FMSLogo 8.4.0.

Refactoring: Rewrite llput to use CAppendableList.
This is shorter and clearer.

Cleanup: Remove an unnecessary special case in lcount() for NIL.
This is already handled by the is_list() case.
I suspect that, at one time, is_list didn't check for NULL before looking at the node type.

Always compile debugutils.cpp, but remove code if DEBUG isn't defined.
This saves the effort of figuring out how to include it when it's needed.

Fix "FD 100 5**6" to run FD 100 before throwing a "not enough inputs to *" error.
This is part of fixing Bug #250.

Fix "FD 100 (" so that it runs FD 100 before throwing an error.

Fix Bug #250; instructions with too many inputs no longer prevent earlier instructions from being run.
This was fixed as a generalization of g_ParseErrorNotEnoughInputs to handle other error types.

There are more parse errors not included in this fix that can prevent earlier instructions from being run.
It's not clear if they'll all be able to use the same mechanism.

Add unnecessary parentheses to stop gcc from warning about if conditions.