Add PDFtk wrapper to /src/pdftk

[tertek]

It could be useful to have a wrapper around php-pdftk that could be residing at /src/Pdftk analog to Filesystem.
Only useful if we can save code duplication though.

[tertek]

/storage
-- /documents/folder_id/document_id <> pdftk free and scan (upload)
-- /outputs/job_id/output_id <> pdftk fill and merge (document, data)
-- /tempDir

// Important: We will work with pdftk-php->getTmpFile (https://github.com/mikehaertl/php-pdftk?tab=readme-ov-file#temporary-file)

// Case create document

$fileUpload = $_FILES['pdf-upload']
$fileUpload['path']

$pdf = new Pdf($path);
$pdf->operation()->execute();
$content = $pdf->getTmpFile();

$file = $this->storage-write($content);

// case create job

$document = $this->storage->read(/documents/$folder_id / $document_id);
// if we cannot give abs path to pdftk_php, we need workaround wiht tmpfile, $tmpFfile = $document
$document_path = $this->storage->getPath($document);
$pdf = new($document_path);

$pdf->fill(..);

$size === 1 only:
$pdf->execute();
$content = $pdf->getTmpFile();

$this->storage->write(/outputs/job_id/output_uid, $content);

[tertek]

@edenst-TPH
I tested the file upload via form, also swapped the PSR7 Implementation and finally found out, that the problem was the VS Code Plugin Rest Client. It seems that some file types cannot be submitted or require a specific syntax that we are not aware of.

I have tested the same requests via Postman which worked fine.
I also created commented on an issue for our currently used Rest Client: Huachao/vscode-restclient#1235

Consequently, we can stay with nyholm/psr7 and test the API with Postman or another more stable client. I hope that the authors of vscode-restclient will fix it soon.

[tertek]

Added tests dae20cf

[tertek]

@edenst-TPH
I have formalized our concept for pdftk & storage from Tuesday.

pdftk & storage concept

Where is pdftk required?

1. Initial actoion to identify form fields of a PDF.
2. Final action of filling and flattening a PDF.

Action Summary

1. Create Document Action:
   User sends multipart post request to create document.

• validate request (required data, existing relations)
• validate document (pdf format, max_limits)
• new pdftk instance from tmp_path of upload, run pdftk --dump_form_fields
• generate uuid
• write to storage
• save to database

2. Create Job Action / Create Output Action:
   User sends field data with Post Request as part of Create Job Action.
   (job creation.. triggers output creation in some kind of queue)

• read document from storage and write into tmp_file
• loop through job, and do for each:
  -- new pdftk instance from tmp_file, run pdftk fill and flatten
  if n>1:
  -- keep as file in array
• zip all files in array
  endif
• generate output uuid
• write to storage as final output
• save/update database (output and job data)

3. Stream Output Action

• Read from Storage
• Stream to Browser as zip (n>1) or pdf (n=1)