There are two issues here. First, you get a warning about the sex check that is part of preprocessQuantile, and second, when you try to plot the predicted sex data, you get an error because you are missing one or more of the required items in your colData object.

The first question to ask yourself is do you really care about that? I mean, maybe you have all female subjects, or maybe there is a mixture, but in the end that might not be something you need to know or care about. If you do care, then you might want to move forward with figuring out what the problem is.

If you do care, do note that estimating the sex is a very simple process. Here is the code

> getSex
function (object = NULL, cutoff = -2) 
{
    .isGenomicOrStop(object)
    if (is(object, "GenomicMethylSet")) 
        CN <- getCN(object)
    if (is(object, "GenomicRatioSet")) 
        CN <- getCN(object)
    xIndex <- which(seqnames(object) == "chrX")
    yIndex <- which(seqnames(object) == "chrY")
    .getSex(CN = CN, xIndex = xIndex, yIndex = yIndex, cutoff = cutoff)
}
<bytecode: 0x000001b852c01b60>
<environment: namespace:minfi>

## the .getSex function is a hidden function that  you can access using the getAnywhere function

> getAnywhere(.getSex)
A single object matching '.getSex' was found
It was found in the following places
  namespace:minfi
with value

function (CN = NULL, xIndex = NULL, yIndex = NULL, cutoff = -2) 
{
    if (is.null(CN) || is.null(xIndex) || is.null(yIndex)) {
        stop("must provide CN, xIndex, and yIndex")
    }
    xMed <- colMedians(CN, rows = xIndex, na.rm = TRUE)
    yMed <- colMedians(CN, rows = yIndex, na.rm = TRUE)
    dd <- yMed - xMed
    k <- kmeans(dd, centers = c(min(dd), max(dd)))
    sex0 <- ifelse(dd < cutoff, "F", "M")
    sex0 <- .checkSex(sex0)
    sex1 <- ifelse(k$cluster == which.min(k$centers), "F", "M")
    sex1 <- .checkSex(sex1)
    if (!identical(sex0, sex1)) {
        warning("An inconsistency was encountered while determining sex. One ", 
            "possibility is that only one sex is present. We recommend ", 
            "further checks, for example with the plotSex function.")
    }
    df <- DataFrame(xMed = xMed, yMed = yMed, predictedSex = sex0)
    rownames(df) <- colnames(CN)
    df
}
<bytecode: 0x000001b852c2e540>
<environment: namespace:minfi>

The basic idea is to get the copy number data, subset that to the X and Y chromosomes, and then use kmeans to cluster into groups. You could do

debug(minfi:::.getSex) 
getSex(rgSet)

## notice that you have to specify the fully qualified function name using the 
## package name followed by three colon characters plus the pre-pended dot
## to debug a function that is hidden in the namespace

And step through the .getSex function and inspect the kmeans cluster values, as well as sex0 and sex1. But again, is this critical for your analysis, and do you have only one sex?