Build .conda artifacts by default

[jaimergp]

Description

Closes #5183

Checklist - did you ...

• Add a file to the news directory (using the template) for the next release's release notes?
• Add / update necessary tests?
• Add / update outdated documentation?

[jakirkham]

Thanks Jaime! 🙏

[codspeed-hq]

CodSpeed Performance Report

Merging #5527 will not alter performance

_{Comparing jaimergp:default-dot-conda (7bad39d) with main (aa7a2b6)}

Summary

✅ 5 untouched benchmarks

[beeankha]

It looks like there is a similar/related issue that was filed here previously: #4890

There's currently an open PR that addresses this issue by creating a new command line argument --package-format and may potentially be the option we want to go with to include in the November 2024 conda-build release. @jaimergp @jakirkham please let me know your thoughts on this.

[jakirkham]

Thanks Bianca! 🙏

Think these are different changes

This change is making the default package type .conda (instead of .tar.bz2) as requested in issue: #5183

That change is allowing users to specify the package type via CLI

These are both good changes. They just serve different purposes. Ideally we get them both in

With this change, we are thinking of having a deprecation (or future) warning in 24.11 to note the default would change in 25.1. Then include this change after the 24.11 branch (or tag)

[jaimergp]

Yea maybe #5209 should go in first (that refactor is useful here too). It's also a good opportunity to add a warning there about the future change in e.g. 25.1. In other words:

For 24.11:

• Review and merge Add command line option to specify the wanted package format #5209
• A new PR adds a warning about the change of the default in 25.1: Warn about upcoming default change in conda_pkg_format #5534

For 25.1:

• Rebase this PR and flip the default. Some tests need adjustment.

[jezdez]

Yea maybe #5209 should go in first (that refactor is useful here too). It's also a good opportunity to add a warning there about the future change in e.g. 25.1. In other words:

For 24.11:

• Review and merge Issue/4890 #5209
• A new PR adds a warning about the change of the default in 25.1

For 25.1:

• Rebase this PR and flip the default. Some tests need adjustment.

I agree with that timeline, just reviewed #5209