Fix an inconsistency with ignorerepeated=true where we weren't

[quinnj]

setting the EOF code after matching repeated delimiters, as noticed by @nickrobinson251 here.

[codecov]

Codecov Report

Base: 90.98% // Head: 90.99% // Increases project coverage by +0.01% 🎉

Coverage data is based on head (9b10767) compared to base (3136584).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #144      +/-   ##
==========================================
+ Coverage   90.98%   90.99%   +0.01%     
==========================================
  Files           9        9              
  Lines        1609     1611       +2     
==========================================
+ Hits         1464     1466       +2     
  Misses        145      145              

Impacted Files	Coverage Δ
src/components.jl	99.17% <100.00%> (+<0.01%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[nickrobinson251]

so now we have:

with ignorerepeated===false

julia> Parsers.codes(Parsers.xparse(String, "a,").code)
"SUCCESS: OK | DELIMITED "

julia> Parsers.codes(Parsers.xparse(String, "a,\n").code)
"SUCCESS: OK | DELIMITED "

with ignorerepeated===true

julia> Parsers.codes(Parsers.xparse(String, "a,,"; ignorerepeated=true).code)
"SUCCESS: OK | DELIMITED | EOF "

julia> Parsers.codes(Parsers.xparse(String, "a,,\n"; ignorerepeated=true).code)
"SUCCESS: OK | DELIMITED | NEWLINE | EOF "

I think this makes the behaviour that's dependent on ignorerepeated be internally consistent, which is an improvement on the status quo
which was

julia> Parsers.codes(Parsers.xparse(String, "a,,"; ignorerepeated=true).code)
"SUCCESS: OK | DELIMITED "

julia> Parsers.codes(Parsers.xparse(String, "a,,\n"; ignorerepeated=true).code)
"SUCCESS: OK | DELIMITED | NEWLINE | EOF "

so on that basis i approve...

but i'm still not sure i'm fully convinced that behaviour should be so different between ignorerepeated===false and ignorerepeated===true.

since CSV.jl depends on it being this way, and it might be the right way even though i don't fully understand it, i think we can leave this as-is and potentially review again another time