rq 0.0.7 is now released and available for installation. The biggest new
feature is fake data generation tools based on jaswdr/faker
(https://github.com/jaswdr/faker).

* bump embedded OPA version from v0.54.0 to v0.56.0
* new builtins:
	* `rq.fake()`
	* `rq.sfake()`
	* `rq.quote()`
	* `rq.unquote()`
* new template functions:
	* `fake`
	* `sfake`
	* `quote`

I added a new unit test to verify the correct behavior:

https://git.sr.ht/~charles/rq/commit/835c5959647bd183f7b44354f8f040c626533420

The intended functionality is that everything after the parameter name becomes the parameter value. Based on the observed behavior with the test, it seems like the first part must be getting chopped off somewhere else. I would suspect it's getting treated as a DataSpec with the format name being `/dev/null # TODO(sr)` and the file path being ` switch to output-format null when`. It's presumably defaulting to the JSON handler.

My rationale for avoiding any extra smarts, such as allowing comments on directive lines was to avoid needing to worry about string escaping issues. Say someone wanted to pass a `#` character to a directive for some reason (say perhaps a field separator for `raw`).

I think the right thing to do for now is to just document this behavior clearly on the scripts page (which I will do presently), but I'm willing to discuss further if you feel strongly that comments should be allowed in directives. 

~ Charles