Add almost-infinite scenario with Clark2Dt dispersal

[juntyr]

• Implement Clark 2Dt dispersal kernel
  • Standard dispersal
  • Dispersal to a different location
    • implemented with rejection sampling
  • Dispersal to a different location probability
    • implemented with numerical integration
    • Does normal dispersal need a correction?
    • Does noise-filtered normal dispersal need a correction?
  • Enable circle origin sampling with arbitrary centres
  • Enable both circle and rectangle origin sampling for both almost-infinite scenario variants
• Fix CI failures from CUDA install
• Squash commits and add @NicoFirbas as a co-author

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 37.06294% with 450 lines in your changes are missing coverage. Please review.

Project coverage is 16.35%. Comparing base (8c6e722) to head (e95edaa).
Report is 12 commits behind head on main.

Files	Patch %	Lines
...cogs/dispersal_sampler/almost_infinite_clark2dt.rs	46.11%	222 Missing ⚠️
rustcoalescence/src/args/config/scenario.rs	0.00%	45 Missing ⚠️
necsim/impls/cuda/src/cogs/maths.rs	0.00%	33 Missing ⚠️
...cogs/dispersal_sampler/in_memory/cumulative/mod.rs	0.00%	27 Missing ⚠️
...std/src/cogs/origin_sampler/singleton_demes/mod.rs	0.00%	25 Missing ⚠️
.../src/cogs/origin_sampler/singleton_demes/circle.rs	0.00%	24 Missing ⚠️
.../src/cogs/dispersal_sampler/in_memory/alias/mod.rs	0.00%	9 Missing ⚠️
...gs/dispersal_sampler/in_memory/packed_alias/mod.rs	0.00%	9 Missing ⚠️
...dispersal_sampler/in_memory/separable_alias/mod.rs	0.00%	9 Missing ⚠️
...ispersal_sampler/in_memory/cumulative/dispersal.rs	0.00%	5 Missing ⚠️
... and 14 more

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #273      +/-   ##
==========================================
+ Coverage   15.19%   16.35%   +1.16%     
==========================================
  Files         315      297      -18     
  Lines       21401    21116     -285     
==========================================
+ Hits         3251     3453     +202     
+ Misses      18150    17663     -487     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[juntyr]

@NicoFirbas I've now added support for p != 1 and the scenario config now takes a tail_p parameter

[juntyr]

@NicoFirbas

Can we add a note stating that the 2Dt tends to Gaussian as p becomes large and tends to Cauchy as p goes to zero? Also could be have the default value for the p parameter be 1?

Done :)

[juntyr]

@NicoFirbas I've now started merging the Clark2Dt scenario into the AlmostInfinite scenario (though it is not quite done yet - only rectangle sampling areas are supported thus far). You now need to enable the almost-infinite-clark2dt-dispersal-scenario and use the following scenario config:

scenario: AlmostInfinite(
    sample: Rectangle(x: 0, y: 0, width: 10000, height: 10000),
    dispersal: Clark2Dt(shape_u: 150, tail_p: 1.0),
)

[juntyr]

Small fixes are needed, then it should be good to go

[juntyr]

@NicoFirbas The config format has slightly changed again as I've done a final pass over the code changes - it's now:

scenario: AlmostInfinite(
    sample: Rectangle(origin: (x: 0, y: 0), width: 10000, height: 10000),
    dispersal: Clark2Dt(shape_u: 150, tail_p: 1.0),
)

[juntyr]

@NicoFirbas I tried to add you as a co-author of this PR but it seems I used the wrong email address - could you share one that is linked to your GitHub account so that I can fix the commit authorship?

[NicoFirbas]

Hi Juniper. That is strange. To my knowledge the email I have linked to my github account is this one ***@***.***). It is where I am receiving all of my email notifications as well. Nicolás Firbas https://www.researchgate.net/profile/Nicolas_Firbas

…

On Sat, Mar 2, 2024 at 11:28 AM Juniper Tyree ***@***.***> wrote: @NicoFirbas <https://github.com/NicoFirbas> I tried to add you as a co-author of this PR but it seems I used the wrong email address - could you share one that is linked to your GitHub account so that I can fix the commit authorship? — Reply to this email directly, view it on GitHub <#273 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATJMXQFBJ6JAMZF5GXJNKYDYWGSWLAVCNFSM6AAAAABDP55LB2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZUG42TONRZGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>