Improve handling of chapter-less contexts

[renehamburger]

There are two issues with the current implementation, which can be seen in

blinx.js/src/blinx.class.spec.ts

Lines 241 to 261 in 72d8184

	it('works for reference without chapter (initial implementation)', async () => {
	await testRecognition(
	`<p data-bx-context="Matt">
	Check out verse 9 and then 6:10 (cf. Luke 11:2)
	and verse 11.
	</p>`,
	['verse 9', '6:10', 'Luke 11:2', 'verse 11'],
	['Matt.1.9', 'Matt.6.10', 'Luke.11.2', 'Matt.1.11']
	);
	});
	xit('works for reference without chapter (final implementation)', async () => {
	await testRecognition(
	`<p data-bx-context="Matt">
	Check out verse 9 and then 6:10 (cf. Luke 11:2)
	and verse 11.
	</p>`,
	['6:10', 'Luke 11:2', 'verse 11'],
	['Matt.6.10', 'Luke.11.2', 'Matt.6.11']
	);
	});

As the given context does not contain a chapter:

• chapter-less references within the section should take their chapter from any previously recognised passage in that section (rather than the chapter 1 supplied to the chapter-less context to allow parsing)
• if a chapter-less reference has no preceding recognised passage, it should be skipped