src/spicelib/parser/inpptree.c, bug fix, B expression of kind `max(0,expr)'

identified and reported by Marcel Hendrix,
expressions of this kind could trigger a segmentation violation.

PTdifferentiate() roughly evaluates to
  ternary_fcn(ge0(0-expr), 0, PTdifferentate(expr))
and mkb() optimizes
  0 - expr --> unary_minus(expr)

IFeval() invokes PTeval() for the derivative too,
PTeval() looks at the incorrect tree->funcnum
  and tries to PTeval for a second argument which is not there,
     (unary_minus does not have a second argument)
  causing a segmentation fault.

src/spicelib/parser/inpptree.c

@@ -827,7 +827,7 @@ static INPparseNode *mkf(int type, INPparseNode * arg)
 
     p->left = inc_usage(arg);
 
-    p->funcnum = i;
+    p->funcnum = funcs[i].number;
     p->function = funcs[i].funcptr;
     p->funcname = funcs[i].name;
 

tests/regression/parser/bxpressn-1.cir

@@ -507,6 +507,18 @@ v1151_g  n1151_g 0  '1'
 v1152_g  n1152_g 0  '1'
 v1153_g  n1153_g 0  '0'
 
+* ----------------------------------------
+* special regression test unveils ...
+*   max(a,b)
+
+vaux1 n42 0 dc 42.0
+
+b1154_t  n1154_t 0  v = max(0, v(n42))
+b1155_t  n1155_t 0  v = max(v(n42), 0)
+
+v1154_g  n1154_g 0  '42.0'
+v1155_g  n1155_g 0  '42.0'
+
 .control
 
 define mismatch(a,b,err) abs(a-b)>err
@@ -516,7 +528,7 @@ op
 let total_count = 0
 let fail_count = 0
 
-let tests = 1001 + vector(153)
+let tests = 1001 + vector(155)
 
 foreach n $&tests
   set n_test = "n{$n}_t"