Replace CaseType's own XML encoding function

[artfulrobot]

The problem is detailed at https://lab.civicrm.org/dev/core/-/issues/1719

In a nutshell:

1. Creating a case type crashes if the name of a case status is not valid XML content. e.g. if a name has & in it, but other special characters would also cause problems.

2. It dies because it fails to properly encode the value.

3. It uses its own encoding functions instead of a proper XML encoder.

This PR:

1. replaces the homespun XML encoder with php's own built in one, and properly encodes everything.

2. adds a test fixture for the case that & is in a status name.

As a side effect, I believe it fixes 1719

[civibot]

(Standard links)

• If this is your first pull-request for CiviCRM, please browse CONTRIBUTING.md for information about the development and testing processes.
• If you are reviewing this pull-request, you may wish to consult the test sites and the Review Standards (long template, short template).

[artfulrobot]

Note: I had to update the xml fixtures because PHP's https://www.php.net/manual/en/book.xmlwriter.php writes the doctype a little differently (UTF not utf and no space before the closing ?>).

[mattwire]

ping @demeritcowboy

[demeritcowboy]

The test failure is real but seems like it's mostly just the utf->UTF thing. There's also a whitespace difference between what came out before and after, but that seems mostly clerical. If anybody is relying on specific whitespace in their xml they shouldn't be, but the test is complaining.

[artfulrobot]

Thanks @demeritcowboy Pushed new commit with fixture updates (whtespace and case of 'utf' → 'UTF' only) seems to get the test passing. Tested locally for the failed test and the original tests, all fine, but let's see what Jenkins has to say about it.

[colemanw]

Code is an excellent improvement, and test coverage is solid.