Revert "Parser: restore the old syntax for lambda-bindings in logical terms"

This reverts commit b19660c9.
We prefer not having two syntaxes for the same thing.
(With the approval of Andrei, of course.)

src/parser/lexer.mll

@@ -187,8 +187,6 @@ rule token = parse
       { AND }
   | "\\/"
       { OR }
-  | "\\"
-      { LAMBDA }
   | "."
       { DOT }
   | ".."

src/parser/parser.mly

@@ -121,7 +121,7 @@
 %token AND ARROW
 %token BAR
 %token COLON COMMA
-%token DOT DOTDOT EQUAL LAMBDA LTGT
+%token DOT DOTDOT EQUAL LTGT
 %token LEFTPAR LEFTPAR_STAR_RIGHTPAR LEFTSQ
 %token LARROW LRARROW OR
 %token RIGHTPAR RIGHTSQ
@@ -153,7 +153,7 @@
 %nonassoc LARROW
 %nonassoc RIGHTSQ    (* stronger than <- for e1[e2 <- e3] *)
 %left OP2
-%left OP3 LAMBDA
+%left OP3
 %left OP4
 %nonassoc prec_prefix_op
 %nonassoc LEFTSQ
@@ -587,7 +587,6 @@ triggers:
 quant:
 | FORALL  { Dterm.DTforall }
 | EXISTS  { Dterm.DTexists }
-| LAMBDA  { Dterm.DTlambda }
 
 numeral:
 | INTEGER { Number.ConstInt $1 }
@@ -907,7 +906,6 @@ lident_op:
 | op_symbol UNDERSCORE    { prefix $1 }
 | EQUAL                   { infix "=" }
 | OPPREF                  { prefix $1 }
-| LAMBDA                  { infix "\\" }
 | LEFTSQ RIGHTSQ          { mixfix "[]" }
 | LEFTSQ LARROW RIGHTSQ   { mixfix "[<-]" }
 | LEFTSQ RIGHTSQ LARROW   { mixfix "[]<-" }
@@ -936,7 +934,6 @@ prefix_op:
 | o = OP2   { mk_id (infix o)    $startpos $endpos }
 | o = OP3   { mk_id (infix o)    $startpos $endpos }
 | o = OP4   { mk_id (infix o)    $startpos $endpos }
-| LAMBDA    { mk_id (infix "\\") $startpos $endpos }
 
 (* Qualified idents *)