feat: Add Intercom Tickets stream support with incremental replication

[mporracindie]

Description of change

This PR adds support for the Intercom Tickets API to tap-intercom, enabling extraction of ticket data using incremental replication with the Search API.

Key Changes:

• Added new tickets stream using POST /tickets/search endpoint
• Implemented incremental replication with updated_at bookmark
• Supports cursor-based pagination with starting_after parameter
• Extracts all ticket fields including custom attributes, contacts, ticket parts, and linked objects
• Follows existing pattern from Contacts and Conversations streams

Files Modified:

• tap_intercom/schemas/tickets.json (NEW) - JSON schema with 20+ ticket fields
• tap_intercom/streams.py - Added Tickets class (55 lines) and registered in STREAMS dict
• README.md - Updated documentation with tickets stream details
• CHANGELOG.md - Added version 2.3.0 entry

Technical Details:

• Replication method: INCREMENTAL (query filtered)
• Primary key: id
• Replication key: updated_at
• Foreign keys: contact_ids, admin_assignee_id, team_assignee_id
• Pagination: 150 records per page with intermediate bookmarks
• Custom attributes supported via additionalProperties: true

Manual QA steps

• Run discovery mode and verify tickets stream appears in catalog

  tap-intercom --config config.json --discover | grep "tickets"                                                                                              

• Run full sync and verify tickets are extracted
  tap-intercom --config config.json --catalog catalog.json
• Verify incremental sync respects bookmark (run twice with state file)
• Verify interrupted sync resumes correctly using intermediate bookmarks
• Verify all ticket fields are extracted (id, title, description, contacts, ticket_parts, etc.)
• Verify timestamps are converted to ISO 8601 format
• Verify custom ticket_attributes are extracted
• Verify pagination works correctly

Test Results:

• Successfully extracted 450+ tickets from test workspace
• All fields properly formatted and timestamps converted
• Pagination working with cursor-based navigation
• Bookmarks written correctly after each page

Risks

• Low Risk: This is a new stream addition with no changes to existing functionality
• No modifications to existing streams, client, or transform logic
• Follows well-established patterns from Contacts and Conversations streams
• No breaking changes to existing API contracts or schemas
• Mitigation: Tickets stream is opt-in via catalog selection; existing streams unaffected

Rollback steps

• Revert this branch
• Alternative: Remove tickets from catalog selection in production until issues are resolved
• No database migrations or state changes required for rollback

AI generated code

https://internal.qlik.dev/general/ways-of-working/code-reviews/#guidelines-for-ai-generated-code

• [ x ] this PR has been written with the help of GitHub Copilot or another generative AI tool