Entity mesh fields should be validated before database insertion to prevent batch failures

Problem/Motivation

The entity_mesh module does not validate field values before inserting them into the database. When processing batch operations with malformed link data, the module attempts to insert oversized values into database columns, causing INSERT queries to fail with truncation errors.

Currently, when an INSERT fails due to data validation issues (e.g., string data too long for a column), the batch process halts with a database error rather than gracefully handling the failed item. This prevents the batch operation from completing and makes it difficult to identify which specific items caused the failure.

Error encountered:

SQLSTATE[22001]: String data, right truncated: 1406 Data too long for column 'target_host' at row 4

In this case, the target_host column received a full URL or text content instead of just the hostname portion, exceeding the column's character limit.

Steps to reproduce

1. Create a content entity (e.g., node) with render links containing malformed or oversized data
2. Configure entity_mesh to track links for this entity type
3. Trigger a batch operation to process entity_mesh data (e.g., via cron or manual batch execution)
4. Include a link where:
   • target_host contains a full sentence or paragraph instead of just the hostname (e.g., "This is a very long description that should not be in the hostname field")
   • Or any field contains data that exceeds its database column size constraint
5. The batch process will fail with a database truncation error and stop processing

Example data structure that causes the issue:

INSERT INTO "entity_mesh" (..., "target_host", ...) VALUES 
(..., "example.com", ...), 
(..., "cdn.example.org", ...), 
(..., "www.example.net", ...), 
(..., "This is a very long text description that exceeds the column size limit", ...)

Proposed resolution

Implement validation for all entity_mesh fields before database insertion:

1. Add field validation logic that checks:
   • String field lengths against their database column constraints
   • Required fields are populated
   • URL components (scheme, host, path) are correctly formatted
   • Entity references point to valid entities
2. Handle validation failures gracefully:
   • Catch validation errors before INSERT attempts
   • Log failed items with detailed error information
   • Mark failed items in the batch tracker so they can be:
     • Retried later
     • Reported to administrators
     • Excluded from causing batch failure
3. Allow batch processing to continue even when individual items fail validation, similar to how robust batch operations handle partial failures