Add note to docs about sharp memory usage on default Linux setup

[lennym]

It's fairly opaque to the user that there are problems with sharp in an out-of-the-box Linux environment, and so if sharp is going to be explicitly recommended for non-Vercel production environments then I think it's a good idea to make sure people are aware of the possible issues.

My vanilla Heroku deployment was using wild amounts of memory until I found the details in the sharp docs about how to configure the memory allocator, so signposting this a bit better should help prevent issues in future.

Bug

• Related issues linked using fixes #number
• Integration tests added
• Errors have helpful link attached, see contributing.md

Feature

• Implements an existing feature request or RFC. Make sure the feature request has been accepted for implementation before opening a PR.
• Related issues linked using fixes #number
• Integration tests added
• Documentation added
• Telemetry added. In case of a feature if it's used or not.
• Errors have helpful link attached, see contributing.md

Documentation / Examples

• Make sure the linting passes by running yarn lint

[molebox]

Thank you!