feat(scanner): match library files by series name and position

[vavallee]

Summary

• Adds Series and SeriesNumber fields to ParsedFile
• ParseFilename now extracts series info from bracket notation ([Series Name #N], [Series Name, Book 1]) and paren notation ((Series Name #N), (Series Name, Book 1)) — series name must start with a letter to avoid false-positives on ISBNs and years
• Falls back to parent directory name for ABS-style layouts where the folder carries the series+number (Author/Series Name, Book 1/title.m4b)
• ScanLibrary falls back to series+position matching when ASIN and title+author both fail — exact series-title match + position, returns no match when ambiguous (two series with the same name both have a book at that position)
• Adds WithSeriesRepo builder to Scanner
• New GetBookBySeriesPosition repo method in SeriesRepo

Fixes library scan failures for users migrating from LazyLibrarian

Users whose existing libraries have series-based filenames like:

[Dune Chronicles #1] Dune - Frank Herbert.m4b
[Stormlight Archive, Book 1] The Way of Kings - Brandon Sanderson.epub
(Dune Chronicles, Book 1) Dune - Frank Herbert.m4b

These previously returned unmatched on library scan. With this PR they reconcile correctly against wanted books via series name + position.

Notes

• ABS layout Author/Series Name/01 - Title.m4b (no number in folder name) still relies on tag reading or title match — the parser can't infer series number from a plain folder name without a position
• ABS layout Author/Series Name, Book 1/Title.m4b (number in folder name via paren) is fully supported

Test plan

• go test ./internal/importer/... ./internal/db/...

[codecov]

Codecov Report

❌ Patch coverage is 12.82051% with 68 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
internal/db/series.go	0.00%	41 Missing ⚠️
internal/importer/scanner.go	0.00%	16 Missing and 1 partial ⚠️
internal/importer/parser.go	50.00%	7 Missing and 3 partials ⚠️

📢 Thoughts on this report? Let us know!