.name_repair in as_tibble.matrix() deprecated

[cyklee]

To Reproduce

uwotplot <- umap_pq(ps, pkg = "uwot",
    n_neighbors = 5,
    verbose = TRUE)

Console output

Taxa are now in rows.
Joining with `by = join_by(Sample)`
23:12:06 UMAP embedding parameters a = 1.577 b = 0.8951
23:12:06 Read 23 rows and found 9714 numeric columns
23:12:06 Using FNN for neighbor search, n_neighbors = 5
23:12:07 Commencing smooth kNN distance calibration using 6 threads with target n_neighbors = 5
23:12:08 Initializing from normalized Laplacian + noise (using RSpectra)
23:12:08 Range-scaling initial input columns to 0-10
23:12:08 Commencing optimization for 500 epochs, with 150 positive edges using 6 threads
23:12:08 Using rng type: pcg
Using method 'umap'
Optimizing with Adam alpha = 1 beta1 = 0.5 beta2 = 0.9 eps = 1e-07
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
23:12:09 Optimization finished
Joining with `by = join_by(Sample)`
Warning message:
The `x` argument of `as_tibble.matrix()` must have unique column names if `.name_repair` is
omitted as of tibble 2.0.0.
ℹ Using compatibility `.name_repair`.
ℹ The deprecated feature was likely used in the MiscMetabar package.

  Please report the issue at <[https://github.com/adrientaudiere/MiscMetabar/issues](vscode-file://vscode-app/Applications/Positron.app/Contents/Resources/app/out/vs/code/electron-browser/workbench/workbench.html#)>.

It appears tibble since 2.0.0 wants users to explicitly use the .name_repair argument to handle non-unique or missing column names. It's probably fine unless they fully defunct this function.

Expected behavior
Complete the process without warning.