747: update identifier validation message now that leading colons aren't allowed

[lindsay-stevens]

Closes #747

Why is this the best possible solution? Were any other approaches considered?

The majority of these commits are organising error/warning messages into the errors.py module, which is what I had in mind for the ErrorCode pattern which was added to pyxform a few months ago. After this PR about 30 error messages are organised there, but pyxform raises an error at about 170 locations and many of those have distinct messages.

The goal of this stage of organisation is to try and identify all the errors/warnings which deal with identifiers, so that the validation approach and message content can be more consistent with each other and with the documentation. Also to take the opportunity to move over error/warning messages that are already using the errors.Detail class or seemed easy to move. There are still some name-related errors that have not been moved (e.g. Audits must always be named 'audit.', etc), but they don't specifically deal with identifier grammar. There is also still a name-like regex check for the "app" parameter but that requires an android package name which has different rules to XML names.

What are the regression risks?

If library or script usages were looking for specific error strings then these changes would break them. But putting the errors into an enum/object should make it easier to avoid looking for specific strings - as is shown in the tests that use ErrorCode.

Does this change require updates to documentation? If so, please file an issue here and include the link below.

Maybe? #630 says ODK tools don't support having a colon as the first character in identifier names but that doesn't seem to be stated in the ODK XForms Specification. On one hand maybe it is not a hard requirement from a spec perspective, but on the other hand it is functionally required via pyxform validation and not being supported by ODK form clients.

Before submitting this PR, please make sure you have:

• included test cases for core behavior and edge cases in tests
• run python -m unittest and verified all tests pass
• run ruff format pyxform tests and ruff check pyxform tests to lint code
• verified that any code or assets from external sources are properly credited in comments

[lindsay-stevens]

Realised there's a name setting so added check/tests for that. It's previously untested, only documented on xlsform.org, and kind of sneaks in there in via json_dict.update(settings) on xls2json.py L352.