Fix smoothed image computation when mask is None in canny

[rfezzani]

Description

Fixes #6346

Checklist

• Docstrings for all functions
• Gallery example in ./doc/examples (new features only)
• Benchmark in ./benchmarks, if your changes aren't covered by an
  existing benchmark
• Unit tests
• Clean style in the spirit of PEP8
• Descriptive commit messages (see below)

For reviewers

• Check that the PR title is short, concise, and will make sense 1 year
  later.
• Check that new functions are imported in corresponding __init__.py.
• Check that new features, API changes, and deprecations are mentioned in
  doc/release/release_dev.rst.
• There is a bot to help automate backporting a PR to an older branch. For
  example, to backport to v0.19.x after merging, add the following in a PR
  comment: @meeseeksdev backport to v0.19.x
• To run benchmarks on a PR, add the run-benchmark label. To rerun, the label
  can be removed and then added again. The benchmark output can be checked in
  the "Actions" tab.

[rfezzani]

@stefanv, the bug #6346 you reported is due to a change in image preprocessing when no mask is provided. In this case the mask is all ones and the constant mode introduces artifacts in the border of the image as cval=0 by default.
Should we change default mode (as already proposed #4342 (comment)) in this PR or in another one?

[grlee77]

Thanks, @rfezzani. Let's do the change to canny's mode default in a separate PR.

[grlee77]

@meeseeksdev backport to v0.19.x