Adding another way of calculating gradients in downsamplers

[XianzheMa]

This PR addresses https://github.com/eth-easl/modyn/issues/505. A flag is introduced, to switch between using purely the last layer gradients, or the concatenation of last layer gradients and penultimate layer gradients, in selection algorithms.

[github-actions[bot]]

^{( % to main) ( % to main)}

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 91.54930% with 6 lines in your changes missing coverage. Please review.

Project coverage is 82.83%. Comparing base (70ce04c) to head (84c20ff).

Files	Patch %	Lines
...sampling_strategies/craig_downsampling_strategy.py	0.00%	2 Missing :warning:
...ling_strategies/gradmatch_downsampling_strategy.py	0.00%	2 Missing :warning:
...ing_strategies/submodular_downsampling_strategy.py	0.00%	2 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #540 +/- ## ========================================== - Coverage 82.84% 82.83% -0.01% ========================================== Files 220 220 Lines 10202 10232 +30 ========================================== + Hits 8452 8476 +24 - Misses 1750 1756 +6 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[MaxiBoether]

Thanks, the new tests look good. Feel free to merge!