Previously, the method directly converted a byte array to a string
without accounting for a potential UTF-8 Byte Order Mark (BOM).
This commit introduces logic to remove the BOM (if present)
using `TrimStart('\uFEFF')` after converting the byte array
to a string.
Additionally, a comment was added to clarify the purpose of
this change.
Added `PdfALevel` and `PdfAWarning` properties to the `PdfExtractionResult` class to store the PDF/A compliance level and indicate if a warning should be displayed for non-compliance.
Updated `Upload.cshtml` to display the PDF/A compliance level and conditionally show a warning message if the document is not PDF/A compliant.
Enhanced `PdfAttachmentExtractorService` to determine the PDF/A compliance level using `PdfDocumentProcessor`, map it to a string representation, and log the compliance level. Added logic to set `PdfAWarning` for non-compliant documents.
Enhanced `ZugferdInvoice` model with default string values to prevent nulls. Updated `Upload.cshtml` to display parsed invoice data.
Refactored `Upload.cshtml.cs` to handle ZUGFeRD XML parsing and database storage. Introduced `ImportedInvoice` property and buffered file processing with `MemoryStream`.
Extended `ZugferdParserService` to support ZUGFeRD v1, v1.0 FeRD, and v2/Factur-X. Added version-specific parsing methods and namespaces. Improved date and decimal parsing for robustness.
Added database migration (`20260522084606_InitialCreate`) to define `ZugferdInvoices` table. Updated migration snapshot to reflect schema changes.
Fixed localization issue in `Upload.cshtml.cs` error message.
