feat: add rich-indexer which is another built-in indexer based on relational database

[EthanYuan]

What problem does this PR solve?

Problem Summary:

CKB already has a built-in indexer, which is based on a key-value database for indexing. Rich-Indexer is a complement to the original indexer, as it is based on a relational database, providing support for more complex and flexible queries.

Currently, Rich-Indexer supports two types of database drivers.

1. One is the embedded SQLite, which users can start using without any configuration.
2. The other is PostgreSQL, which can undergo independent, customized configurations for both software and hardware. Additionally, it allows users to engage in secondary development based on it.

What is changed and how it works?

What's Changed:

Once the Rich-Indexer is activated, the CKB node's RPC based on the Rich-Indexer will gain additional capabilities.

INDEXER RPC	Indexer	Rich-Indexer
get_cells script args partial mode search	❌	✔️
get_cells cell data filter(prefix|exact|partial)	✔️	✔️
get_transactions script args partial mode search	❌	✔️
get_transactions cell data filter(prefix|exact|partial)	❌	✔️
get_cells_capacity script args partial mode search	❌	✔️
get_cells_capacity cell data filter(prefix|exact|partial)	✔️	✔️

In addition, Rich-Indexer can also expose full block data in the form of a PostgreSQL database, providing the most basic and flexible indexing format accessible via SQL and compatible with any programming language.

Please refer to the README for more detailed information

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No code ci-runs-only: [ quick_checks,linters ]

Side effects

• Performance regression

Release note

Title Only: Include only the PR title in the release note.