feat: update processing of AAR file/codes

[PsypherPunk]

Context

The codes used for AAR were updated some time ago: all future AAR files can be expected to adhere to the newer BAE-/BAI-prefixed codes (similarly BTE/BTI for AAR CS data).

AB#241700

Change proposed in this pull request

• previous BNCH… codes have been updated with BAI…/BAE… codes
• update processing of AAR file to accommodate the above, in line with recent changes for input_schemas
• update processing of AAR/CS file to accommodate the above, in line with recent changes for input_schemas

Guidance to review

Have verified that for all years 2021—2023, the files produced by this updated code produce the same files as exist in t01:

❯ md5sum --check ../../t01/2023/checksum.txt
aar.parquet: OK
academies.parquet: OK
all_schools.parquet: OK
bfr_metrics.parquet: OK
bfr.parquet: OK
cdc.parquet: OK
census.parquet: OK
central_services.parquet: OK
cfo.parquet: OK
ks2.parquet: OK
ks4.parquet: OK
maintained_schools.parquet: OK
schools.parquet: OK
sen.parquet: OK
trusts.parquet: OK

I've also run the data for 2024 locally, with 2 documented discrepancies:

• all Trust names are now upper-case (purely a cosmetic issue to be noted)
• the CS data is missing some Trusts referenced in the AAR data (again, to be noted but the data are incomplete until later this month)

Checklist (add/remove as appropriate)

• Work items have been linked (use AB#)
• Your code builds clean without any errors or warnings
• You have run all unit/integration tests and they pass
• Your branch has been rebased onto main
• You have tested by running locally
• You have reviewed with UX/Design

[jrabbott]

LTGM - would be good to get @benlav-50 eyes over the mappings.

Might also be worth adding some documentation on the schema changes?