[core] Call useGridColumnPinning before useGridColumns

[flaviendelangle]

Each hook using GridPreProcessingGroup.hydrateColumns and called after useGridColumns causes a full additional pre-processing on mount because the columns need to be pre-processed before and after the 1st call of the feature hook.
Whereas if the feature hook is before, the pre-processing will only run once when useGridColumns is called.

The same applies to other pre-processings but this one is especially heavy when we have a lot of columns and has a lot of hooks hooked to it, so we could quickly run 5 or more useless pre-processings if we don't take care.

To more we move toward event listeners and pre-processors, the more useGridRows and useGridColumns should be late in the feature hook call order.

Same things as described here #3387 (comment)

[mui-bot]

These are the results for the performance tests:

Test case	Unit	Min	Max	Median	Mean	σ
Filter 100k rows	ms	190.6	297.7	232	243.12	43.14
Sort 100k rows	ms	385	656.9	611.5	579.42	99.42
Select 100k rows	ms	198.8	277.2	242.4	240.08	28.152
Deselect 100k rows	ms	125.4	261.1	202.9	194.32	45.117

Generated by 🚫 dangerJS against 58efc03