Add detection of strings in code sections with stricter threshold

Currently we only detect strings in fileinfo (with `--strings` option) in what is assumed as sections containing data. However there are cases when there might be strings also in sections containing code or sections that are for some reason not marked as a data sections. The original premise of this was to lower the amount of garbage detected as strings (as some byte sequences might represent readable strings).

However not having certain strings detected can be punishing, so we should also start detecting strings in all sections while making the threshold stricter for sections which do not resemble data sections. Threshold would be the same for data sections as it was until now.

Start at fileinfo's `--strings` handling and trace how sections are classified as data or code. Review the existing string-detection threshold and its section filtering. Done means strings are considered in all sections, with the existing threshold for data sections and a stricter threshold for other sections.

Written by the indexing model from the issue text.