#2973 letsimp treats constant factor 1/2 differently than 2

[Robert Dodier]

In %o4 I expected to see foo(m, n)/2, likewise in %o6 I expected foo(m, n)/7.

(%i1) matchdeclare ([a, c], true, b, lambda ([e], is (e # 1))) $
(%i2) let (c*a^b, c*foo(a, b));
                              b
(%o2)                        a  c --> foo(a, b) c
(%i3) letsimp (2*m^n);
(%o3)                             2 foo(m, n)
(%i4) letsimp ((1/2)*m^n);
                                       n
                                      m
(%o4)                                 --
                                      2
(%i5) letsimp ((3/7)*m^n);
                                  3 foo(m, n)
(%o5)                             -----------
                                       7
(%i6) letsimp ((1/7)*m^n);
                                       n
                                      m
(%o6)                                 --
                                      7

Maxima version: "5.36.1"
Maxima build date: "2015-04-30 13:41:12"
Host type: "i686-pc-linux-gnu"
Lisp implementation type: "CLISP"
Lisp implementation version: "2.49 (2010-07-07) (built 3605610186) (memory 3639415274)"