tests: re-enable 2086, and 472, 1299, 1613 for Windows

[vszakats]

• GHA/windows: un-ignore tests 472 1299 1613.
  They were ignored for the mingw-w64 c-ares U job.
  They do run fine now:
  https://github.com/curl/curl/actions/runs/12032875421/job/33547724780?pr=15644

• globally re-enable test 2086.
  Comment says it only affected Windows.
  Seems to be running fine now. Example:
  Windows:
  https://github.com/curl/curl/actions/runs/12032875421/job/33547718309?pr=15644#step:13:3856
  Linux:
  https://github.com/curl/curl/actions/runs/12032875397/job/33545739712#step:41:3650

• update comments for disabled tests 1184, 1801.
  They affect all operating systems, likely all CIs.

  FAIL 1801: 'HTTP/2 upgrade with lying server' HTTP, HTTP/2
  https://github.com/curl/curl/actions/runs/12032362497/job/33544053001#step:14:4265
  FAIL 1184: 'HTTP 1.1 CONNECT with redirect and set -H user-agent' HTTP, HTTP CONNECT, HTTP proxy, proxytunnel
  https://github.com/curl/curl/actions/runs/12032362497/job/33544051415#step:14:5252

[MarcelRaad]

Ah, test 2086 [Pre-request callback for HTTP IPv6] still fails for me:
https://curl.se/dev/log.cgi?id=20241128050932-2865694#prob1
https://curl.se/dev/log.cgi?id=20241128062717-2916001#prob1
Because of:

=== Start of file stderr2086
URL: [;1];60179\2086#ipv6

Looks like we have to tell MSYS2 to leave what it recognizes as path separators alone.

[vszakats]

Ah, test 2086 [Pre-request callback for HTTP IPv6] still fails for me: https://curl.se/dev/log.cgi?id=20241128050932-2865694#prob1 https://curl.se/dev/log.cgi?id=20241128062717-2916001#prob1 Because of:

=== Start of file stderr2086
URL: [;1];60179\2086#ipv6

Looks like we have to tell MSYS2 to leave what it recognizes as path separators alone.

Isn't this a function of MSYS2 version? Its changelog says something changed around here recently:
https://www.msys2.org/news/#2024-11-03-disabling-mingw-w64-wildcard-support-by-default

If so, I still have no idea how to detect this and force it to work for older MSYS2 versions.

[MarcelRaad]

I think we just need to set MSYS2_ARG_CONV_EXCL in <setenv> as done in other tests, which is unrelated to globbing IIUC. No idea yet what the difference between my machine and the CI builds is.

[vszakats]

I think we just need to set MSYS2_ARG_CONV_EXCL in <setenv> as done in other tests, which is unrelated to globbing IIUC. No idea yet what the difference between my machine and the CI builds is.

Are you also using the freshest MSYS2?

edit: based on buildinfo.compiler.version: 10.2.0 via https://curl.se/dev/log.cgi?id=20241203050535-3961987, this may be an older MSYS2 without the recent change.

That envvar is suspect indeed, though I had very little experience/success with it,
and my only way to try is in CI.

We can certainly put it back to the DISABLE list if there is no obvious solution.

edit: a milder solution is to ignore this test via TFLAGS in the failing envs.

[MarcelRaad]

Yes, it's indeed an older version - I'm a bit hesitant to update it as updates regularly break the e-mail client. I can confirm that setting MSYS2_ARG_CONV_EXCL as in other tests fixes the issue for me. Maybe newer versions just have a better path detection.
PR here: #15677