feat: add NDJSON and YAML exporters to ScrapeResult + extension-aware save()

[sharyaropensource]

Fixes #145

Summary

Adds line-oriented / config-friendly exporters to ScrapeResult and extends save() to infer them from file extensions.

New methods

to_ndjson() -> str

• Emits one JSON object per line over data (the NDJSON contract, no metadata/errors envelope)
• Uses ensure_ascii=False so unicode is preserved literally
• Empty data -> "" (matches to_csv empty guard)

to_yaml() -> str

• Dumps the full {data, metadata, errors} envelope as YAML
• Guarded by optional PyYAML import using the same ImportError pattern as to_dataframe
• Install with pip install pyscrappy[yaml]

Extended save() mappings

Extension	Method
.ndjson, .jsonl	to_ndjson()
.yaml, .yml	to_yaml()
.json, .csv, .md	unchanged

Changes

• src/pyscrappy/core/models.py: Added to_ndjson(), to_yaml(), updated save() and class docstring
• tests/test_core/test_models.py: 13 new tests covering NDJSON round-trip, unicode, empty data, YAML round-trip, missing PyYAML error, and all new save() extensions

Test plan

• All 33 tests pass (20 existing + 13 new)
• NDJSON: round-trip parse each line, unicode preserved, empty data -> empty string
• YAML: round-trip via yaml.safe_load, missing PyYAML raises ImportError
• save(): .ndjson/.jsonl/.yaml/.yml all write correctly