Feature request: sequential pattern mining

Sequential pattern mining is a general data mining solution for finding patterns in sequences e.g. if you have a text something like
```
ABCD
CBCD
BECD
```
then it can find the `BCD` as a frequent closed sequence with the support of 3. It is good to prepare the data first, e.g. in the case of logs split up to words and give each word an individual id and each special character an individual id, use the algorithm on the resulting array and do reverse id -> word mapping on the results. This is necessary to spare CPU and memory, otherwise it would eat a lot of resources. Different type of data might need different preparation. SPM can be used for objects with different properties too and the algorithms can check multiple properties, not just a single one, so in the case of JSON, not plain text this can be an extra feature.

I don't think there is any alternative tool for this. There are countless algorithms for this, I don't know which would be the best, definitely not the old ones like GSP or SPADE. They might need a different thread, otherwise it might freeze the window.

The issue names no files, tests, or entry points; start by reviewing CyberChef's existing operation structure and how data-analysis features are tested. Before implementation, define the sequential pattern mining algorithm, supported input and output formats, resource limits, and acceptance tests.

Written by the indexing model from the issue text.