Fix to default colors in phenogram

https://raw.githubusercontent.com/liamrevell/phytools/master/R/phenogram.R

A phytools reader reported the following issue in using the phytools traitgram function, phenogram, to plot a tree with a mapped discrete character:

library(phytools)

## Loading required package: ape
## Loading required package: maps

tree<-pbtree(n=20,scale=2)
Q=matrix(c(-1,1,1,-1),2,2)
rownames(Q)<-colnames(Q)<-LETTERS[1:2]
x1<-sim.history(tree,Q,anc="A")$states

## Done simulation(s).

x1

## t14 t15  t9  t1 t19 t20 t18 t10 t12 t13 t11 t16 t17  t4  t2  t3  t5  t6 
## "A" "B" "B" "A" "A" "A" "A" "A" "A" "A" "A" "A" "A" "A" "A" "B" "A" "A" 
##  t7  t8 
## "A" "A"

y<-fastBM(tree)
t1<-make.simmap(tree,x1)

## make.simmap is sampling character histories conditioned on the transition matrix
## Q =

##            A          B
## A -0.2959494  0.2959494
## B  0.2959494 -0.2959494

## (estimated using likelihood);
## and (mean) root node prior probabilities
## pi =

##   A   B 
## 0.5 0.5

## Done.

phenogram(t1,y)

By contrast, if we use a numerically indexed discrete character, we don't face the same issue:

x2<-sapply(x1,function(x) if(x=="A") "1" else "2")
t2<-make.simmap(tree,x2)

## make.simmap is sampling character histories conditioned on the transition matrix
## Q =

##            1          2
## 1 -0.2959494  0.2959494
## 2  0.2959494 -0.2959494

## (estimated using likelihood);
## and (mean) root node prior probabilities
## pi =

##   1   2 
## 0.5 0.5

## Done.

phenogram(t2,y)

The issue is not that deep. In fact, it can be circumvented if we are willing to just use the argument colors (as we would in plotSimmap) to set the colors of our mapped discrete trait. E.g.,

colors<-setNames(c("blue","red"),LETTERS[1:2])
colors

##      A      B 
## "blue"  "red"

phenogram(t1,y,colors=colors)

I have nonetheless fixed this issue so that we will get the mapped discrete state colored by default if no colors are specified.

source("https://raw.githubusercontent.com/liamrevell/phytools/master/R/phenogram.R")
phenogram(t1,y)

phytools is now on GitHub, so in fact this update can be installed automatically in a fresh R session using the devtools package as follows:

## don't run
library(devtools)
install_github("liamrevell/phytools")

That's all folks.