Describe the bug
I think the IDL functions lowest_set_bit and highest_set_bit are defined slightly incorrectly.

In reviewing #471, the following assignment was being slightly corrected:

X[rd] = (xlen() - 1) - $signed(lowest_set_bit(X[rs1]));

• X[rd] and X[rs1] are of type Bits[XLEN] (I believe, as BUILTIN).
• xlen() is defined:

  function xlen {
    returns Bits<8>
  

  (I'll note that Extension B: fix ctz instruction IDL code #471 is removing the use of xlen(), but it's useful for additional context here.)
• I confess not knowing where $signed is defined (?)
• lowest_set_bit(...) is defined:

  function highest_set_bit {
    returns XReg
    arguments XReg value
  [...]
  function lowest_set_bit {
    returns XReg
    arguments XReg value
  

It makes sense to me that the *set_bit functions accept an XReg parameter.

It does not make sense to me that they return an XReg value. Ignoring $signed, the calculation is using register values to compute an integer value (less than or equal to XLEN).

Should these functions instead return Bits<8> (or essentially Bits<log2(XLEN)>, but we presume that 8 is sufficient)?