docs(streams): add detailed explanation of Primitive Streams (IntStream, LongStream, DoubleStream)

What
- Documented **Primitive Streams in Java** and their role in numerical data processing.
- Covered:
  - Definition and difference from object-based streams.
  - Specialized types: IntStream, LongStream, DoubleStream.
  - Built-in numeric methods (sum, average, min, max, summaryStatistics).
  - Benefits: performance optimization, readability, specialized numeric APIs.
- Explained real-world applications across back-end systems, analytics, financial apps, and middleware.
- Provided example using IntStream.rangeClosed(1, 100).sum().
- Added notes on integration with other APIs (boxed conversion, Collectors aggregation).

Why
- Standard `Stream<Integer>` incurs autoboxing/unboxing overhead, reducing efficiency in numeric-heavy workloads.
- Primitive Streams solve this by working directly with primitives (int, long, double).
- They are critical in performance-sensitive domains such as analytics, finance, and real-time monitoring.
- Understanding their role ensures developers can write cleaner, faster, and more maintainable numeric pipelines.

Logic
1. **Specialized Stream Types**
   - `IntStream` → operations on int values.
   - `LongStream` → operations on long values.
   - `DoubleStream` → operations on double values.

2. **Performance Advantage**
   - Avoids autoboxing/unboxing (`int ↔ Integer`).
   - Reduces memory usage and GC pressure.
   - Better suited for large-scale numeric operations.

3. **Built-in Numeric Methods**
   - Directly provides `sum()`, `average()`, `min()`, `max()`, and `summaryStatistics()`.
   - Eliminates need for manual reduction/aggregation logic.

4. **Integration**
   - `boxed()` → converts primitive streams to object streams when APIs require `Stream<T>`.
   - Works seamlessly with Collectors for statistical reporting.

Real-world applications
- **Back-End Processing**: Sales aggregation, DB report generation.
- **Financial Systems**: Risk models, moving averages, high-frequency trading data.
- **Analytics/Scientific Computing**: Sensor data, simulations, dashboards.
- **Middleware/Utilities**: Reusable numeric helper APIs, parallel data crunching.

Notes
- Prefer primitive streams for numeric-heavy workloads.
- Use `parallel()` when large datasets benefit from multicore processing.
- Convert to boxed streams only when integration requires object types.
- Leverage `summaryStatistics()` for concise statistical reporting.

Signed-off-by: https://github.com/Someshdiwan <[email protected]>

Author: Someshdiwan