RFC: Should template mode be removed?

My original goal for Lily was to have a template language. The idea was to have Lily code between <?lily ... ?> tags, and render content outside of the tags as-is. Two of Lily's libraries in garden (apache and postgres) were written with template mode in mind.

While my initial focus was to create a templating language, that focus has shifted toward general use capabilities and, more specifically, embedding. As a result, the apache and postgres libraries have seen very little movement, aside from the occasional update to sync to currently Lily. Both are very feature barren, and currently serve as little more than examples of how to create a library.

I am thinking about removing template mode from the language. I am opening this issue up in case that anyone is using it. If you are using it, I would like to know so that I can keep the feature. I will leave this open until the next release (10/10/25). If it is not being used, I will remove the relevant parts at the beginning of the next release.

I'm not looking to remove any other core features of the language. If this is to be removed, I'll put apache into read-only mode, and likely do the same for postgres as well.

The code for implementing template mode is a small part of the interpreter, and relatively stable as well. This is primarily about making the focus of the language clear, as the code for implementing template mode in the interpreter is relatively small.