Check `earliest_available_slot` when handling by root/range requests

[nflaig]

This was highlighted in a security report

Although it's not a big issue but we should be checking earliest_available_slot and cancel requests early.

This also applies for block requests for post-fulu.

Do we wanna close the connection or throw an error?

[nflaig]

Also related to that, we should be checking start slot against MIN_EPOCHS_FOR_DATA_COLUMN_SIDECARS_REQUESTS in our by range request handler. There isn't much point of looking the data columns up in the DB as they have been pruned.

We do have --chain.archiveDataEpochs flag to extend this period but it's not necessary to serve this data via req/resp.

From the spec

Let `data_column_serve_range` be
`[max(current_epoch - MIN_EPOCHS_FOR_DATA_COLUMN_SIDECARS_REQUESTS, FULU_FORK_EPOCH), current_epoch]`.
Clients MUST keep a record of data column sidecars seen on the epoch range
`data_column_serve_range` where `current_epoch` is defined by the current
wall-clock time, and clients MUST support serving requests of data columns on
this range.