refactor: move processing out of StatementBatch constructor

[boscorat]

Parent issue

#194 — function purification

Problem

StatementBatch.__init__ (statements.py:1278) is a classic "constructor does too much" anti-pattern. On construction it:

• Calls validate_or_initialise_project() which creates directories and DB files
• Calls asyncio.run() / ProcessPoolExecutor for parallel processing
• Processes all PDFs (writing temp parquet files to disk)
• Prints timing summaries

You cannot create a StatementBatch object without actually processing PDFs. There is no way to construct one with mock data for testing.

Proposal

Make the constructor a pure data holder. Move processing into a separate process() or run() method. The constructor should store configuration; the method should execute the pipeline.

Code relationships

Target:

• StatementBatch.__init__() — statements.py:1278

Callers (will need updating):

• StatementBatch.__init__ in conftest.py (test fixture)
• StatementBatch.__init__ in CLI handler (cli.py)

Related (called from init):

• validate_or_initialise_project() — paths.py
• process_pdf_statement() — statements.py:828
• StatementBatch.__process_batch_sequential() — statements.py:1377
• StatementBatch.__process_batch_turbo() — statements.py:1398

Existing tests:

• tests/test_statements.py::TestGoodStatements — creates StatementBatch in conftest fixture
• tests/conftest.py — session fixtures create StatementBatch objects

All existing tests will need updating to call batch.process() after construction.

Siblings

• refactor: BatchMetadata dataclass to replace 13 duplicated parameters #195 — BatchMetadata dataclass
• refactor: ExtractionContext dataclass for extraction pipeline #196 — ExtractionContext dataclass
• refactor: decompose Statement constructor into pure factory + persistence #197 — Statement constructor decomposition
• refactor: decompose process_pdf_statement into pure extraction + persistence #198 — process_pdf_statement decomposition
• refactor: stop mutating inputs in extraction pipeline #199 — Stop mutating inputs
• refactor: add optional conn parameter to build_datamart and related functions #200 — DB connection threading
• refactor: remove dead logs and file_path parameters from pipeline functions #202 — Dead parameter cleanup
• refactor: Parquet class accepts pre-loaded data instead of reading from disk #203 — Parquet class purification
• refactor: _handle_parquet_write_error returns error dict instead of mutating #204 — _handle_parquet_write_error purification