feat(aws_s3): add local_file_path configuration for uploading files

[juniorrhis1]

closes #3925

This pull request adds support for uploading files to AWS S3 from a specified local file path, in addition to the existing functionality of uploading message content. The changes include updates to both the documentation and the implementation to introduce a new configuration field, local_file_path, which allows users to specify a local file to upload per message.

Key changes:

Feature: Support for uploading from a local file path

• Added a new configuration field local_file_path to the S3 output component, allowing users to specify the path of a local file to upload instead of message content. This field supports interpolation functions for dynamic file path resolution. [1] [2] [3] [4] [5] [6]
• Updated the S3 writer logic to check if local_file_path is set for each message; if so, it reads the specified file and uploads its contents, otherwise it uploads the message content as before. [1] [2]
• Improved documentation in aws_s3.adoc to describe the new local_file_path field, its type, default value, and usage examples.

Implementation details

• Refactored the upload logic by introducing a helper method getUploadBody to encapsulate the logic for choosing between local file content and message content.
• Added necessary imports for file handling in Go (os, io).

[CLAassistant]

All committers have signed the CLA.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[mihaitodor]

Hi @juniorrhis1 👋 Could you please let us know why you'd prefer this approach over something like the config below?

input:
  file:
    paths:
      - test.txt
    scanner:
      to_the_end: {}

output:
  aws_s3:
    ...

[juniorrhis1]

Hi @mihaitodor, to avoid load the entire file in memory.

For small files or json files with one document per line (using lines scanner) this approach is prefered, but is not possible send large files without consumes a lot of memory.