Update to nlm 2.9.2 v2

[jamesvillarrubia]

Description of the change

• This PR updates the Tika server version from 2.9.2_v1 to 2.9.2_v2 and adds new headers to the TikaFileParser.
• New headers X-Tika-PDFOcrStrategy and X-Tika-PDFExtractFontNames have been added to the TikaFileParser for better control of the OCR and PDF parsing behavior.
• 2.9.2_v1 failed to included necessary style attributes when computing <p> tags. New jar resolves this bug.

Type of change

• [X] Bug fix (non-breaking change that fixes an issue)
• [ ] New feature (non-breaking change that adds functionality)
• [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)

Testing

The changes have been tested manually to ensure the Tika server works correctly with the new configuration and headers. Additional testing is required for edge cases.

[jamesvillarrubia]

@ansukla As expected, there are some critical fixes in this version that I missed in the first, if you get a sec to a approve. Cheers!

[JSv4]

FWIW, I've been testing some PDFs that are broken under current main branch. Three of them are working great with these changes.

[bzhr]

Can this get published?

[bzhr]

Very nice, if somebody can provide an example of a docker build command that will build from the main branch, so that I can test, that would be wonderful

[Ebraheem-Alrabeea]

Hi @jamesvillarrubia,

I hope you're doing well! I wanted to ask if the new v2 contains any changes other than those listed in the comparison you posted here. It seems unclear what the changes are inside the v2 Jar. I plan to make some RTL fixes over NLM 2.9.2 and would appreciate your confirmation. Also, what is the Apache Tika version or commit you have used?

Thank you!