Add OpenAI-powered content parsing improvement feature

[paskal]

Resolves#27.

[coveralls]

Pull Request Test Coverage Report for Build 11280720445

Details

• 66 of 195(33.85%) changed or added relevant lines in 6 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-12.5%) to 76.539%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
backend/extractor/openai_mock.go	13	24	54.17%
backend/rest/server.go	19	49	38.78%
backend/extractor/readability.go	28	116	24.14%

Totals
Change from base Build 11121542707:	-12.5%
Covered Lines:	659
Relevant Lines:	861

---

💛 - Coveralls

[coveralls]

Pull Request Test Coverage Report for Build 11280720229

Details

• 66 of 195(33.85%) changed or added relevant lines in 6 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-12.5%) to 76.539%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
backend/extractor/openai_mock.go	13	24	54.17%
backend/rest/server.go	19	49	38.78%
backend/extractor/readability.go	28	116	24.14%

Totals
Change from base Build 11121542707:	-12.5%
Covered Lines:	659
Relevant Lines:	861

---

💛 - Coveralls

[paskal]

Superseded by #74, which implements this functionality with a better design: automatic inline evaluation loop instead of a manual endpoint, fail-open semantics, and iterative selector refinement.