Implement new private_api/taxa/shared_peptides endpoint

[pverscha]

Add private_api/taxa/shared_peptides endpoint

shared_peptides

New endpoint POST /private_api/taxa/shared_peptides that accepts a list of taxon IDs at species or strain level, digests all proteins for each taxon, and returns the intersection — peptides present in every taxon after digestion. No index lookup is performed; the result is a pure set intersection across the digested proteomes. Protein retrieval for all taxa runs concurrently.

Parameters

• taxon_ids — list of taxon IDs; each must be at species or strain rank (400 otherwise)
• cleavage_regex — regex defining cleavage sites (default: [KR](?!P), tryptic digestion)
• min_length — minimum peptide length to retain (default: 5)

Response

{
  "shared_peptides": ["PEPTIDE", ...]
}

---

Supporting changes

• database::get_proteins_for_taxon — new function that retrieves all UniprotEntry records for a given taxon ID using search_after pagination, handling taxa with more than 10,000
  proteins correctly.
• TaxonStore::get_rank — new accessor following the existing get_name pattern.
• api/src/helpers/digestion — shared module for cleave_sequence, compile_cleavage_regex, validate_taxon_rank, and serde defaults; placed in helpers/ alongside lca_helper,
  ec_helper, etc. to keep controller files free of domain logic.
• ApiError::InvalidRegexError — new error variant (HTTP 400) for invalid cleavage regex strings, distinct from the existing UnknownRankError.
• fancy-regex — added as a dependency to support lookahead assertions in cleavage patterns (e.g. (?!P)).

[pverscha]

Merged with #101