Co-authored-by: Matthew Peveler <matt.peveler@gmail.com>

This check can reduce the file size significantly when writing PDFs and there are objects used multiple times, for example images.

If you read- and write a file again, this can already reduce the file size:

```python
from PyPDF2 import PdfReader, PdfWriter

reader = PdfReader("big-old-file.pdf")
writer = PdfWriter()

for page in reader.pages:
    writer.add_page(page)

writer.add_metadata(reader.metadata)

with open("smaller-new-file.pdf", "wb") as fp:
    writer.write(fp)
```

For the following files, the size was reduced significantly:

* https://corpora.tika.apache.org/base/docs/govdocs1/958/958893.pdf : 5.7 MB ➔ 0.8 MB (-4.9 MB / -86%)
* https://corpora.tika.apache.org/base/docs/govdocs1/949/949428.pdf : 6.7 MB ➔ 1.9 MB (-4.8 MB / -72%)
* https://corpora.tika.apache.org/base/docs/govdocs1/911/911140.pdf : 8.8 MB ➔ 4.3 MB (-4.5 MB / -51%)

Co-authored-by: Harry Karvonen <harry.karvonen@onebyte.fi>

See PEP-257 for some guides.

flake8-naming was used to find those spots.

The "rtype" (return type) was removed from the docstrings as it already is in the type annotation in a more readable way + linked with the used type. It's also checked by mypy in the function signature.

Signed-off-by: Matthew Peveler <matt.peveler@gmail.com>

When writing a PDF, set the version to the highest PDF version of the
written ones

Closes #162

Co-authored-by: Matthew Peveler <matt.peveler@gmail.com>

Also in writer.encrypt

Closes #585

New Features (ENH):
-  Add writer.pdf_header property (getter and setter) (#1038)

Performance Improvements (PI):
-  Remove b_ call in FloatObject.write_to_stream (#1044)
-  Check duplicate objects in writer._sweep_indirect_references (#207)

Documentation (DOC):
-  How to surppress exceptions/warnings/log messages (#1037)
-  Remove hyphen from lossless (#1041)
-  Compression of content streams (#1040)
-  Fix inconsistent variable names in add-watermark.md (#1039)
-  File size reduction
-  Add CHANGELOG to the rendered docs (#1023)

Maintenance (MAINT):
-  Handle XML error when reading XmpInformation (#1030)
-  Deduplicate Code / add mutmut config (#1022)

Code Style (STY):
-  Use unnecessary one-line function / class attribute (#1043)
-  Docstring formatting (#1033)

Full Changelog: 2.4.0...2.4.1