fix(jobs): handle long log lines and prevent log fetch timeouts

[eleboucher]

Job pods were sticking around after completion because the controller failed to fetch logs. The "bufio.Scanner: token too long" error occurred when log lines exceeded the default 64KB buffer size. Additionally, log fetching had no timeout, causing the controller to hang indefinitely on large log streams, preventing state transitions.

Solution:

1. Added MOVER_LOG_MAX_SCAN_BUFFER_SIZE env var (default: 1024KB = 1MB)
   to configure scanner buffer size for handling long log lines
2. Added MOVER_LOG_FETCH_TIMEOUT env var (default: 120s) to set a
   timeout for log fetching operations
3. Increased scanner buffer in FilterLogs() based on configuration
4. Added timeout context in updateMoverStatusForJob() to prevent hanging
5. Added comprehensive tests for long line handling and configuration

Changes:

• internal/controller/utils/podlogs.go: Added buffer size and timeout
  configuration, increased scanner buffer, added timeout context
• internal/controller/utils/podlogs_test.go: Added tests for long log
  lines, buffer size configuration, and timeout configuration