feat: migrate data layer to SQLite database and add Education module

[Ayesha-Anwar607]

Summary of Changes

This PR migrates the data storage layer of faker-pk from hardcoded Python lists/dictionaries to a packaged, serverless SQLite database (faker_pk.db), and introduces a new Education module with realistic student profiles and institution filtering.

---

Key Improvements & Features Added

1. SQLite Database Storage Layer

• Data Migration: Migrated all raw datasets (names, locations, job titles, industries, companies, SIM prefixes, castes, sects, banks) to a structured SQLite database.
• Database Schema: Created 11 relational tables (industries, job_titles, companies, locations, names, sim_providers, sim_prefixes, castes, sects, banks, institutions) with proper foreign keys and constraints.
• Auto-Initialization Fallback: Implemented an automatic database recreation check in faker_pk/utils.py so that if faker_pk.db is missing, it automatically regenerates from initialize_db.py.
• Package Distribution: Updated pyproject.toml with setuptools package-data configuration to natively package *.db files inside wheels/tarballs.

2. Enhanced Data Logic & Realistic Relational Queries

• Gender-Aware CNICs: cnic(gender=...) generates CNICs ending with an odd digit for males and an even digit for females.
• Provider-Aware SIM Prefixes: phone_number(provider=...) pulls authentic mobile prefixes matching the specified telecom provider.
• Cohesive Address Layouts: full_address() generates real, city-matched Pakistani postal codes alongside province filtering in city() and province().
• Bank-Specific IBAN Codes: iban(bank=...) generates valid Pakistani IBAN formats utilizing authentic 4-letter bank codes (e.g. MEZN for Meezan Bank).
• Expanded Industry Categories: Added 5 new industries: Art & Entertainment, Politics, Agriculture, Domestic & Personal Services, and Defense & Public Safety.

3. New Education Module (education.py)

• Institution Generator: institution(level=..., city=..., province=...) generates schools, colleges, or universities filtered by level or location.
• Age-Appropriate Student DOBs: student_dob(level=...) generates realistic birth dates matching institution age limits (5–14 for school, 14–18 for college, 18–25 for university).
• Cohesive Student Profiles: student_profile(level=..., province=...) generates a complete dictionary containing consistent gender, CNIC, institution, city, province, and age-appropriate DOB.

---

Test Suite & Verification Results

• Modular Test Architecture: Split education unit tests into single-purpose test files under tests/.
• Relational Verification: Added 17 new unit tests in test_realistic_combinations.py testing CNIC gender alignment, address consistency, IBAN codes, SIM prefixes, and DB auto-initialization.
• Results: All 41 unit tests passed successfully in 0.69s.

============================= 41 passed in 0.69s ==============================