DPL Analysis: rework cursor logic keeping the gsl::span for VLAs

[aalkin]

Replace gsl::span with std::span almost everywhere.

Make sure the cursor table fill can accept span.

[github-actions]

REQUEST FOR PRODUCTION RELEASES:
To request your PR to be included in production software, please add the corresponding labels called "async-" to your PR. Add the labels directly (if you have the permissions) or add a comment of the form (note that labels are separated by a ",")

+async-label <label1>, <label2>, !<label3> ...

This will add <label1> and <label2> and removes <label3>.

The following labels are available
async-2023-pbpb-apass4
async-2023-pp-apass4
async-2024-pp-apass1
async-2022-pp-apass7
async-2024-pp-cpass0
async-2024-PbPb-apass1
async-2024-ppRef-apass1
async-2024-PbPb-apass2
async-2023-PbPb-apass5

[ktf]

Looks good. Shall I test it on hyperloop?

[aalkin]

Let's wait for the rest of CI to complete, the issues in the previous version were specifically with o2_slc9

[aalkin]

@ktf this time it was all-green in CI. I've committed the suggested change and now it is ready for testing on Hyperloop.

[ktf]

Tested on hyperloop (Correlations, histograms). No regressions spotted.

[mhemmer-cern]

I think in the ASoA.h we still have:

gsl::span<const _Type_> inline getIds() const                                                        \
    {                                                                                                    \
      return _Getter_##Ids();                                                                            \
    }                                                                                                    \
                                                                                                         \
    gsl::span<const _Type_> _Getter_##Ids() const                                                        \
    {                                                                                                    \
      return *mColumnIterator;                                                                           \
    } 

Which results in table converters when trying to fill values of the old table into the new table in errors, because the new table is expecting std::span but the old table getters are giving gsl::span.