Add intersection parameter to /lineage-mutations

[flaneuse]

Current behavior:

https://api.outbreak.info/genomics/lineage-mutations?pangolin_lineage=B.1,B.3&frequency=0.9 performs multiple queries, to return any mutations within B.1 with a prevalence >= 0.9, and then any mutations within B.3 with a prevalence >= 0.9:

(irrelevant fields not shown for clarity)

{
'B.1': 
[{'mutation': 'orf8:s84l',
   'lineage': 'B.1',
   'prevalence': 0.9953091104120138,
   ... },
  {'mutation': 's:d614g',
   'lineage': 'B.1',
   'prevalence': 0.9929476557559457,
   ...},
  {'mutation': 'orf1b:p314l',
   'lineage': 'B.1',
   'prevalence': 0.9799956773372397,
   ...}],

 'B.3':
 [{'mutation': 'orf8:s84l',
   'lineage': 'B.3',
   'prevalence': 1.0,
   ...},
  {'mutation': 'orf1a:g392d',
   'lineage': 'B.3',
   'prevalence': 0.9836488812392427,
   ...}]
}

Desired behavior:

• intersection=false should be the default behavior, and should be the current behavior, as shown above.
• With the parameter intersection=true, the prevalence of all mutations which are prevalent at >= frequency should be calculated.
• For instance, https://api.outbreak.info/genomics/lineage-mutations?pangolin_lineage=B.1,B.3&frequency=0.9&intersection=true, this should also include:
  • The prevalence of s:d614g in B.3 (0.0008605851979345956)
  • The prevalence of orf1b:p314l in B.3 (0.0008605851979345956)
  • The prevalence of orf1a:g392d in B.1 (0.00007206110781943087)
• Structure of the returned object should be the same, just with an extra object in 'B.1 and two extra objects in 'B.3'