[Draft] revive pdfetch_ONS()

[JanMarvin]

Apologies for the slightly rough initial submission—it was a bit late yesterday evening when I opened the PR.

I briefly explored the package and managed to get the pdfetch_ONS() example working.

This implementation uses the Search API to retrieve the URI, which can then be used to fetch either the data or the CSV. I opted not to use the CSV directly, as it includes a header format that I didn't want to investigate further. Additionally, the API provides useful metadata such as update status, which might be helpful.

Just to clarify: I’m not a user of this package and don’t work with ONS data, so I don’t have a specific use case for this functionality. I've cleaned up the PR a bit, but I won’t be making further contributions in this area.

pdfetch::pdfetch_ONS(c("LF24","LF2G"), "lms")
#>            LF24  LF2G
#> 1971-02-28 72.2 24219
#> 1971-03-31 72.1 24187
#> 1971-04-30 72.0 24171
#> 1971-05-31 72.0 24168
#> 1971-06-30 71.9 24153
#> 1971-07-31 71.8 24125
#> 1971-08-31 71.7 24085
#> 1971-09-30 71.6 24046
#> 1971-10-31 71.5 24015
#> 1971-11-30 71.5 24017
#>        ...           
#> 2024-06-30 74.7 32088
#> 2024-07-31 75.0 32223
#> 2024-08-31 74.9 32224
#> 2024-09-30 74.9 32231
#> 2024-10-31 74.9 32231
#> 2024-11-30 75.0 32298
#> 2024-12-31 75.0 32342
#> 2025-01-31 75.1 32390
#> 2025-02-28 75.0 32383
#> 2025-03-31 75.1 32430

Closes #20 and #16