Add reader for ascii files

[trossi]

This PR adds parser for ascii files.

See also #31.

A minimal test is also added for reading the following files generated in R:

data = list(1.1, 2L, 3+4i, NA, 'aä')
save(data, file='test_ascii_v2.rda', ascii=TRUE, compress=FALSE, version=2)
save(data, file='test_ascii_v3.rda', ascii=TRUE, compress=FALSE, version=3)
saveRDS(data, file='test_ascii_v2.rds', ascii=TRUE, compress=FALSE, version=2)
saveRDS(data, file='test_ascii_v3.rds', ascii=TRUE, compress=FALSE, version=3)

These files are generated both on Linux and Windows. Windows-R seems to use \r\n line ending for rds files but \n for rda files (tested on R 4.3.2).

@vnmabus Could you review? Do you think that this kind of testing would be sufficient?

[codecov-commenter]

Codecov Report

Attention: 7 lines in your changes are missing coverage. Please review.

Comparison is base (2b927cb) 91.04% compared to head (122f847) 90.08%.
Report is 1 commits behind head on develop.

Files	Patch %	Lines
rdata/parser/_parser.py	75.00%	5 Missing ⚠️
rdata/parser/_ascii.py	94.44%	2 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop      #34      +/-   ##
===========================================
- Coverage    91.04%   90.08%   -0.97%     
===========================================
  Files            7        8       +1     
  Lines         1106     1170      +64     
===========================================
+ Hits          1007     1054      +47     
- Misses          99      116      +17     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[trossi]

@vnmabus I rebased the branch to fix conflicts and fixed some mypy issues found locally. The mypy CI action seems not be working, see a log here.

[vnmabus]

I would rather keep the parameters everywhere (for type-checking). Alternatively, you could type it with an unpacked TypedDict as in PEP 692, but I think it does not worth it in this case.

[trossi]

I added the parameters back (d193450), but without default values so that they are not duplicated from _parser.py here.

[trossi]

Line ending of _parser.py had changed in develop branch, so I rebased this PR again (it seemed clearer approach than resolving merge conflict with the two files with different line endings). Style issues by ruff are fixed also.

[vnmabus]

LGTM

[trossi]

@vnmabus Thanks for reviewing and merging!