[workspace] Upgrade doxygen_internal to latest release 1.15.0

[tyler-yankee]

Closes #23778.

---

This change is 

[tyler-yankee]

+a:@BetsyMcPhail for feature review, please? (This one seems much simpler than the last upgrade.)

Reviewable status: LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits), missing label for release notes (waiting on @tyler-yankee)

[tyler-yankee]

Reviewable status: 1 unresolved discussion, LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits) (waiting on @tyler-yankee)

---

a discussion (no related file):
Working

Low chance this gets reviewed before the pending release anyways, but in case it does, we should probably hold off landing until the release? As-is, the next release is the first to even depend on the 1.14.0 binary.

[jwnimmer-tri]

Reviewable status: 1 unresolved discussion, LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits) (waiting on @tyler-yankee)

---

a discussion (no related file):

Previously, tyler-yankee (Tyler Yankee) wrote…

Working

Low chance this gets reviewed before the pending release anyways, but in case it does, we should probably hold off landing until the release? As-is, the next release is the first to even depend on the 1.14.0 binary.

To the best of my understanding, the release and this are unrelated. So long as the CI docs built passes, it should be safe to land anytime.

[tyler-yankee]

Reviewable status: 1 unresolved discussion, LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits) (waiting on @tyler-yankee)

---

a discussion (no related file):

Previously, jwnimmer-tri (Jeremy Nimmer) wrote…

To the best of my understanding, the release and this are unrelated. So long as the CI docs built passes, it should be safe to land anytime.

Okay, thanks. I wasn't sure, because I know we have the retention rules for the binaries on drake-mirror, and it seems odd to have never released a Drake that depends on a binary sitting up there (even though it did on master for a time)...

[jwnimmer-tri]

Reviewable status: LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits) (waiting on @tyler-yankee)

---

a discussion (no related file):

Previously, tyler-yankee (Tyler Yankee) wrote…

Okay, thanks. I wasn't sure, because I know we have the retention rules for the binaries on drake-mirror, and it seems odd to have never released a Drake that depends on a binary sitting up there (even though it did on master for a time)...

Being on master is a sufficient requirement for needing the archive kept around. We need to be able to git bisect against all commits on master (at least, going back for several years).

[tyler-yankee]

Reviewable status: 1 unresolved discussion, LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits) (waiting on @tyler-yankee)

---

a discussion (no related file):
Working

From f2f: try to actually diff the outputs from master vs. this PR and see if it's a useful validation for this upgrade. (Of course, the last upgrade would have been totally impossible to parse, but this one could theoretically be only incremental.)

[tyler-yankee]

Reviewable status: 1 unresolved discussion, LGTM missing from assignee BetsyMcPhail, needs platform reviewer assigned, needs at least two assigned reviewers, commits need curation (https://drake.mit.edu/reviewable.html#curated-commits) (waiting on @tyler-yankee)

---

a discussion (no related file):

Previously, tyler-yankee (Tyler Yankee) wrote…

Working

From f2f: try to actually diff the outputs from master vs. this PR and see if it's a useful validation for this upgrade. (Of course, the last upgrade would have been totally impossible to parse, but this one could theoretically be only incremental.)

I tried looking into this, and perhaps with some more time I could get to something useful, but this is nontrivial.

My current command is:

diff -r -I '\([Dd]oxygen[[:space:]]\+[0-9]\+\.[0-9]\+\.[0-9]\+\)\|\(mathjax@[34]\)\|\(<\|>\|&lt;\|&gt;\)' doxygen-out-1.14.0/ doxygen-out-1.15.0/ > diff.txt

which accounts for differences like:

• Generated by Doxygen 1.{14|15}.0
• mathjax@{3|4} in URLs
• apparently in at least some cases, they've replaced literal < and > with the HTML variants < >

Of the remaining cases (the diff is still massive), what it seems like on the surface is:

• some generated hashes after a filename
• how the edge indices of the class graphs are generated

There's a case in the operator== overload in formula.h:623 I want to flag here to check later, the diff looks non-trivial.