simulationFrames are requested repeatedly before first one completes

[robnagler]

Again this is NERSC sbatch, and there's a bug where the reply isn't coming back. Yet, the frames are continually requested:

16:24:08 wsreq#36 send: /simulation-frame/1734884626*multiElectronAnimation*M4LvkaPE*srw*HashIsUnused*1734884317*0*0*1 null
sirepo.js?20241222.161745:3 16:24:23 wsreq#37 send: /run-status {computeJobHash: 'HashIsUnused', computeJobStart: 0, computeJobSerial: 1734884317, simulationId: 'M4LvkaPE', simulationType: 'srw', …}
sirepo.js?20241222.161745:3 16:24:23 wsreq#37 reply: {state: 'running', dbUpdateTime: 1734884655, elapsedTime: 0, frameCount: 1734884650, computeJobStart: 0, …}
sirepo.js?20241222.161745:3 16:24:23 wsreq#38 send: /simulation-frame/1734884649*multiElectronAnimation*M4LvkaPE*srw*HashIsUnused*1734884317*0*0*1 null
sirepo.js?20241222.161745:3 16:24:38 wsreq#39 send: /run-status {computeJobHash: 'HashIsUnused', computeJobStart: 0, computeJobSerial: 1734884317, simulationId: 'M4LvkaPE', simulationType: 'srw', …}
sirepo.js?20241222.161745:3 16:24:38 wsreq#39 reply: {state: 'running', dbUpdateTime: 1734884670, elapsedTime: 0, frameCount: 1734884650, computeJobStart: 0, …}
sirepo.js?20241222.161745:3 16:24:53 wsreq#40 send: /run-status {computeJobHash: 'HashIsUnused', computeJobStart: 0, computeJobSerial: 1734884317, simulationId: 'M4LvkaPE', simulationType: 'srw', …}
sirepo.js?20241222.161745:3 16:24:53 wsreq#40 reply: {state: 'running', dbUpdateTime: 1734884685, elapsedTime: 0, frameCount: 1734884676, computeJobStart: 0, …}
sirepo.js?20241222.161745:3 16:24:53 wsreq#41 send: /simulation-frame/1734884675*multiElectronAnimation*M4LvkaPE*srw*HashIsUnused*1734884317*0*0*1 null

@moellep this is another weird case, but it has to do with the decentralized nature of the requests. Maybe this is not easily fixable.