colVars for dgeMatrix

I am working with large sparse matrices that, after some operations, can return non-sparse matrices of class `dgeMatrix`.

I must be missing something here, but when I try to get colVars for a `dgeMatrix`, I get

```
Error in MatrixGenerics:::.load_next_suggested_package_to_search(x) :
Failed to find a colVars() method for dgeMatrix objects.
```

even though `colMeans2` works just fine.

Here is some (hopefully) reproducible code:

```
m <- Matrix(nrow = 3, ncol = 6, data = rnorm(18), sparse = FALSE)
sparseMatrixStats::colVars(m)
```

PS: `matrixStats::colVars(m)` returns

```
Error in matrixStats::colVars(m) :
Argument 'x' must be a matrix or a vector.
```

and calling `as.matrix` prior to `matrixStats::colVars` is very slow on larger matrices.

Any help would be much appreciated.

Start by reproducing the example with sparseMatrixStats::colVars() on a dgeMatrix, then trace the colVars dispatch alongside the working colMeans2 path. The issue is resolved when colVars works for dgeMatrix inputs without converting large matrices with as.matrix().

Scritto dal modello di indicizzazione a partire dal testo della issue.