Emergency Self-Construct

@lowkeyrossi FWIW: our GHAs have been under constant development since the beginning of the year and are still in flux. Note that even for wheels, we do need `conda` to...

> [@esc](https://github.com/esc) I am pretty aware of CIBW wheel builds and it usage. Probably I can experiment it with building latest llvm-lite source for Win-ARM64 in my fork and let...

@lowkeyrossi here is the warning about using `cibuildwheel`: https://github.com/numba/llvmlite/blob/release0.44/.github/workflows/release.yml#L26-L35 ...this code has been removed quite recently and it is unlikely that a wheel build process based on `cibuildwheel` is going...

> [@esc](https://github.com/esc) We’re facing a few challenges in supporting Windows ARM64 conda packages for llvmlite, since conda support for Windows ARM64 is not yet available and may take some time....

> Yes as you said, adopting a conda-free approach would require refactoring the existing CI setup. However, since the timeline for Miniconda support on Windows ARM is uncertain and may...

@Greenie0701 we started to discuss this today, however, most of the meeting time was taken up discussing the current in-progress release. We will chat about this again next week and...

> Regarding the workflow draft: I have been experimenting with multiple workflow drafts. I paused because llvmlite underwent major changes in the past two weeks, and I want to align...

> OK, that sounds agreeable, I'll ping you on this issue as soon as it is a good time. @Greenie0701 just a quick heads up. The Numba 0.62.* / llvmlite...

@Greenie0701 quick update here: we managed to establish a Tier based support policy framework for the Numba project: PR: https://github.com/numba/numba/pull/10283 Rendered doc: https://numba.readthedocs.io/en/latest/reference/support_tiers.html Windows-on-Arm is currently in Tier 2. In...