Refactor configuration.json migration to initial-config profiles

[Aberdur]

PR Checklist

• This comment contains a description of changes (with reason).
• Make sure your code lints (black and flake8).
• CHANGELOG.md is updated.
• README.md is updated (including new tool citations and authors/contributors).

PR Description

This PR completes the configuration migration from configuration.json to project-specific initial configs, and adapts core modules to work with merged runtime config (configuration.json + extra_config.json) per project profile.

Main focus:

• New/updated project initial configs.
• download, read-lab-metadata, and validate alignment with the new config model.
• Cleanup of project-specific values from configuration.json.

Motivation

We now support multiple project profiles (RELECOV, EQA2026, MEPRAM) with different operational settings.
To avoid hardcoded project values in the generic config, project-specific behavior is moved to initial profile YAMLs and then loaded into extra_config.

Key Changes

1) Initial config split and migration

• Added/updated:
  • relecov_tools/conf/initial_config-relecov.yaml
  • relecov_tools/conf/initial_config-EQA2026.yaml
  • relecov_tools/conf/initial_config-MEPRAM.yaml
• Kept configuration.json as generic baseline and removed migrated project-specific blocks.
• Updated initial_config.yaml and profile defaults as needed.

2) download adaptation

• Switched config loading to merged runtime config:
  • ConfigJson() → ConfigJson(extra_config=True).
• Made metadata sample-id column configurable via:
  • sftp_handle.metadata_processing.sample_id_col
  • fallback remains "Sample ID given for sequencing".
• Updated duplicate-sample warning text to use the configured sample-id column name.

3) read-lab-metadata adaptation

• Uses merged config (ConfigJson(extra_config=True)) and resolves project/profile config from extra_config.
• Added profile-level control for schema casting:
  • cast_values_from_schema (project-level override, fallback to global read_lab_metadata config).
• adding_ontology_to_enum now supports ontology enums both from:
  • direct properties.<field>.enum, and
  • $ref-resolved enums (e.g. #/$defs/enums/...).
• adding_fields() now sets:
  • s_json["map_field"] = sample_data_map_field
  • s_json["file"] = "samples_data.json" (prevents missing-key issues in downstream processing).
• Type conversion in metadata parsing is now conditional:
  • only enforced when cast_values_from_schema=true,
  • with adjusted handling for numeric/date-like Excel cells to preserve behavior across projects.

4) validate adaptation

• Column tag for invalid Excel is now derived from schema label:
  • properties[sample_id_field].label
  • fallback "Sample ID given for sequencing".

5) Related module alignment

• Updated other modules that consume migrated config sections:
  • sftp_client.py
  • build_schema.py
  • upload_database.py
  • read_bioinfo_metadata.py
  • map.py
• Added/updated EQA schema artifacts:
  • relecov_tools/schema/relecov_schema_EQA.json
  • updates in relecov_tools/schema/relecov_schema.json

Files of Interest

• Config:
  • relecov_tools/conf/configuration.json
  • relecov_tools/conf/initial_config-relecov.yaml
  • relecov_tools/conf/initial_config-EQA2026.yaml
  • relecov_tools/conf/initial_config-MEPRAM.yaml
• Core modules:
  • relecov_tools/download.py
  • relecov_tools/read_lab_metadata.py
  • relecov_tools/validate.py

[Shettland]

LGTM