Introduction of random access reading

[Borewit]

Introduce Random Access Support via supportsRandomAccess and IRandomAccessTokenizer

This pull request enhances the strtok3 library by adding support for random access reading, a significant improvement for scenarios requiring non-sequential data access. The core of this update is the introduction of:

supportsRandomAccess Method:

This method provides a way to determine whether a given tokenizer supports random access, offering developers clarity and control over tokenization strategies.
IRandomAccessTokenizer Interface:

A new interface type specifically designed for tokenizers that allow random access to data, enabling developers to implement or extend tokenizers with advanced reading capabilities.

Key Benefits

Random Access Reading: Facilitates efficient handling of files where data needs to be read or processed in non-linear sequences.
Enhanced API Design: Introduces a clear distinction between sequential and random-access tokenizers, improving type safety and extensibility.
Flexibility: Enables broader use cases, such as working with large files, network streams, or complex data formats that benefit from selective data access.

Motivation

The addition of random access support aligns with the need for more versatile file-processing capabilities, as seen in related discussions around file-type detection and malformed file handling. This feature is particularly useful for integrating with applications requiring adaptive or performance-optimized reading strategies.

Allow random access to the underlying media, if the underlying media supports it.

Related to sindresorhus/file-type#688

[coveralls]

coverage: 94.726% (-0.2%) from 94.894%
when pulling ccfec12 on feature-random-access-tokenizer
into 8b4f02b on master.

[Borewit]

Part of strtok3 version 9.1.0