Add TCGA (GDC) / (ICGC)=PCAWG

[dtitov]

Sumana's use-case (if it's still relevant).

[dtitov]

GDC and ICGC APIs are not suitable for crawling: there’s no way to download metadata - one can only search using their endpoints.

PCAWG is better: there’s a way to download metadata (http://pancancer.info/gnos_metadata/latest/), however the metadata is too detailed (https://drive.google.com/file/d/1K-wRabGt4pIBJqZTPhmC0KIDfm7C1frM/view?usp=sharing), thus huge (raw files are 900MB of JSON -> 6529004 entries in DB). TrackFind can’t visualize such an amount of data.