Working with data.table

Nobody has claimed this yet.

Great book!
Working with data.table does not always work:

```
pacman::p_load(data.table, regtools, qeML)

data(day1)
setDT(day1)

# page 17 , 1
data(mlb)
setDT(mlb)
w = mlb[, .(Height, Weight, Age)]
z = qeKNN(w, "Weight")
# Warning message:
# In eval(tmp, parent.frame()) : "data" converted to data frame
predict(z, data.table(Height = 72, Age = 24)) # does not work
# Error in `[.data.table`(dfr, , i) :
# j (the 2nd argument inside [...]) is a single symbol but column name 'i' is not found. Perhaps you intended DT[, ..i]. This difference to data.frame is deliberate and explained in FAQ 1.1.

# page 17 , 2
y = mlb[, 3:6]
knnout1 = qeKNN(y, "Weight", k = 25)
# Warning message:
# In eval(tmp, parent.frame()) : "data" converted to data frame
predict(knnout1, data.table(Height = 72, Age = 24, Position = "Catcher")) # works!
```

In view of the much better performance of data.table on larger datasets I'd rather avoid data.frames and tibbles.

```
> sessionInfo()
R version 4.2.1 (2022-06-23)
Platform: aarch64-apple-darwin20 (64-bit)
Running under: macOS 14.1

Matrix products: default
LAPACK: /Library/Frameworks/R.framework/Versions/4.2-arm64/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] qeML_1.1 tufte_0.13 rmarkdown_2.25 regtools_1.7.0 gtools_3.9.4 FNN_1.1.3.2 data.table_1.14.8

loaded via a namespace (and not attached):
[1] Rcpp_1.0.11 compiler_4.2.1 partools_1.1.6 iterators_1.0.14 tools_4.2.1 rpart_4.1.21 digest_0.6.33 evaluate_0.22 lattice_0.22-5 rlang_1.1.1 Matrix_1.5-4.1 foreach_1.5.2
[13] cli_3.6.1 mlapi_0.1.1 rstudioapi_0.15.0 parallel_4.2.1 RhpcBLASctl_0.23-42 gbm_2.1.8.1 mvtnorm_1.2-3 xfun_0.40 fastmap_1.1.1 knitr_1.45 text2vec_0.6.3 pdist_1.2.1
[25] glmnet_4.1-8 grid_4.2.1 rje_1.12.1 grf_2.3.1 R6_2.5.1 FOCI_0.1.3 survival_3.5-7 pacman_0.5.1 carData_3.0-5 lgr_0.4.4 car_3.1-2 codetools_0.2-19
[37] htmltools_0.5.6.1 MASS_7.3-60 splines_4.2.1 rpart.plot_3.1.1 abind_1.4-5 float_0.3-1 shape_1.4.6 rsparse_0.5.1 sandwich_3.0-2 crayon_1.5.2 zoo_1.8-12
```

Start by reproducing the qeKNN and predict examples in the issue with data.table inputs, then trace the qeKNN and predict entry points to see where the input is converted or indexed. Done means the documented examples work with data.table inputs without the reported warning or error, with behavior covered by an appropriate regression test.

Written by the indexing model from the issue text.